Growth, Yield and Chemical Composition of Pelargonium Sidoides DC

Growth, yield and chemical composition of Pelargonium sidoides DC. in response to nitrogen and soil water management

Motiki Meshack Mofokeng

Submitted in partial fulfilment of the requirements for the degree M.Inst.Agrar. (Agronomy)

Department of Plant Production and Soil Science In the Faculty of Natural and Agricultural Sciences University of Pretoria Pretoria

Supervisor: Prof. J.M. Steyn Co-Supervisor: Dr. C.P. du Plooy Co-Supervisor: Dr. G. Prinsloo

April 2015

UNIVERSITY OF PRETORIA

DECLARATION OF ORIGINALITY

Full names of student: MOTIKI MESHACK MOFOKENG

Student number: 10602209

Declaration

I, Motiki Meshack Mofokeng, declare that the dissertation, which I hereby submit for the degree MIsntAgrar: Agronomy is my own original work and has not previously been submitted by me for a degree at this or any other tertiary institution.

SIGNATURE STUDENT:………………………………………………………………………

SIGNATURE SUPERVISOR:………………………………………………………………………

ACKNOWLEDGEMENTS

For all the achievements in my life and studies “Glory be to the Almighty”.

It gives me pleasure to acknowledge many individuals and organizations whom this study would not have been complete without their encouragement and contribution.

My sincere thanks and deepest gratitude goes to my supervisor Prof. J.M. Steyn and co- supervisors Dr C.P. du Plooy and Dr G. Prinsloo. Their support, patient guidance, encouragement and excellent advice throughout my studies are highly appreciated.

My grateful acknowledgment goes to the Department of Science and Technology, South Africa, for funding this study.

Many thanks go to all my colleagues, including N.R. Sathekge, K.A. Thaoge, T.S. Ramphisa, K.M. Nyathi, Dr Y. Beletse and Dr H.T. Araya for their support in different times of my study. Many thanks to A. Nkadimeng, S.O. Mailula, A.N. Maleka, C.R. Baloyi, G.K. Sethole and Z. Makhubela for their assistance in planting, maintaining and harvesting the trial. Mrs Itumeleng Setshedi (CSIR) is also thanked for her assistance with NMR analysis.

I am thankful to my wife Monicah, our son Toka and daughter Hali for all the love, encouragement and for allowing me to steal their time for my studies. This work is dedicated to my late parents Pheello Samson and Mamphakiseng Martha Mofokeng; and my late brother Tsoabetla Abram Mofokeng. I have not forgotten that no matter our background, you always believed that with education we can achieve what we aspire to. To my sisters Kgoabane, Mahadi and Mapaseka, let us “keep the candle burning forever”.

“It does not matter where you are coming from, what matters is where you are going. So, work hard and shine on”

TABLE OF CONTENTS

PAGE

LIST OF TABLES vi LIST OF FIGURES vii APPENDICES x ABSTRACT xi

CHAPTERS

1 GENERAL INTRODUCTION 1 1.1. Background 1 1.2. African medicinal plant industry 1 1.3. Rationale for study 2 1.3.1. Study purpose and objectives 3 1.3.2. Hypotheses 4 1.3.3. Structure of dissertation 4 References 5 2 LITERATURE REVIEW 8 2.1. Introduction 8 2.2. Pelargonium sidoides 8 2.2.1. Botany and distribution 9 2.2.2. Bio-activity and uses of P. sidoides 12 2.2.3. Wild harvest of P. sidoides 13 2.2.4. Propagation and cultivation 14 References 17 3 PHYSIO-MORPHOLOGICAL RESPONSE OF P. SIDOIDES DC. TO SOIL WATER AND NITROGEN LEVEL 2 2 Abstract 22 3.1. Introduction 23 3.2. Materials and methods 25 3.2.1. Study area 25 3.2.2. Trial design, plant material and duration 26

iii

3.2.3. Treatment application 26 3.2.4. Agronomic practices 28 3.2.5. Data collection and statistical analysis 28 3.2.6. Scanning electron microscope (SEM) observation 29 3.3. Results and discussion 29 3.3.1. Growth response 29 3.3.2. Stomatal conductance 34 3.4. Conclusions 39 References 40 4 YIELD RESPONSE OF P. SIDOIDES DC. TO SOIL WATER AND NITROGEN MANAGEMENT 46 Abstract 46 4.1. Introduction 47 4.2. Materials and methods 49 4.2.1. Data collection and statistical analysis 49 4.3. Results and discussion 50 4.3.1. Soil water deficit and water use 50 4.3.2. Total fresh biomass yield 51 4.3.3. Root yield 53 4.3.4. Chlorophyll and nitrogen content 54

4.3.5. Nitrogen uptake and nitrogen use efficiency (NUEroot) 56

4.3.6. Water use efficiency (WUEroot) 58 4.4. Conclusions 59 References 61 5 CHEMICAL COMPOSITION OF P. SIDOIDES DC. AS INFLUENCED BY NITROGEN AND SOIL WATER LEVEL 68 Abstract 68 5.1. Introduction 70 5.2. Materials and methods 72 5.2.1. Thin layer chromatography (TLC) 72 5.2.2. Metabolomic analysis 72 5.2.2.1. Sample preparation and extraction method 72 5.2.2.2. Data acquisition/sample analysis 72

5.2.2.3. Data mining and processing 73 5.3. Results 73 5.4. Discussion 81 5.5. Conclusions 82 References 83 6 GENERAL DISCUSSION, CONCLUSIONS AND RECOMMENDATIONS 87

LIST OF TABLES

TABLE PAGE

3.1. Physical properties of the soil at the experimental site per 20 cm depth increment to a depth of 100 cm. 25 3.2. Chemical properties of the soil at the experimental site. 25 3.3. Summary of weather data collected during the experiment period. 26 3.4. Averages of plant height, number of leaves and leaf area index of P. sidoides over the growth period. 30 3.5. Average leaf area index, leaf area per plant and stomatal conductance of P. sidoides, in response to water treatment. 33 4.1. Fresh, dry root yield and total biomass dry matter content (percentage) of P. sidoides in response to water stress. 54 4.2. Nitrogen content (%) in the leaves and in whole P. sidoides plant in response to water treatment. 55 5.1. Specific NMR regions and compounds which contributed to the separation of the well watered treatment from the water stressed treatment. 79

LIST OF FIGURES FIGURE PAGE

2.1. Dark maroon flowers (A) of P. sidoides as compared to the pink flowers of P. reniforme (D). The leaves of P. sidoides (B) and those of P. reniforme (E). Tuberous roots of P. sidoides (C) with their internal reddish colour (F). 10 2.2. Geographical distribution of P. sidoides (maroon circles) and P. reniforme (pink squares) in South Africa. Source: SANBI distribution data. 11 2.3. Some of the pharmaceutical products, made from root extracts of P. sidoides that are available over the counter. Source: www.nativa.co.za 13 3.1. Leaf area index of P. sidoides in response to water treatment, over the growing period. LSD: least significant difference, ADL: allowable depletion level. 30 3.2. Plant height of P. sidoides over the growing period in response to water treatment. LSD: least significant difference, ADL: allowable depletion level. 31 3.3. Number of leaves of P. sidoides over the growing period, in response to water treatment. LSD: least significant difference, ADL: allowable depletion level. 32 3.4. Average number of leaves per plant, of P. sidoides, in response to nitrogen level, at harvesting. LSD: least significant difference. 34 3.5. Average stomatal conductance of P. sidoides at different times over the growing period, in response to water stress. LSD: least significant difference, ADL: allowable depletion level. 35 3.6.A. P. sidoides leaf samples showing stomatal pores as observed under an electron microscope. O = old leaves, M = mature leaves and Y = young leaves. B = abaxial side and T = adaxial side. 37 3.6.B. P. sidoides leaf samples showing stomatal pores as observed under an electron microscope. O = old leaves, M = mature leaves and Y = young leaves. B = abaxial side and T = adaxial side. 38

vii

4.1. Changes in soil water deficits (top 400 mm root zone) for the well watered (30% ADL), moderately stressed (50% ADL) and severely stressed (70% ADL) treatment. 51 4.2. Total fresh biomass yield of P. sidoides in response to different nitrogen application levels. 52 4.3. The effect of soil water level on total fresh biomass yield of P. sidoides. 52 4.4. Chlorophyll content (SPAD) of P. sidoides as affected by soil water level. 55 4.5. Nitrogen content, on dry mass basis, of whole plants of P. sidoides as affected by nitrogen level. 56 4.6. Total nitrogen uptake, on a dry mass basis, of P. sidoides plants exposed to different water levels. 57 4.7. Nitrogen use efficiency of P. sidoides plants, on a dry root mass basis, as influenced by nitrogen levels. 58 4.8. Nitrogen use efficiency of P. sidoides, on a dry root mass basis, as influenced by soil water level. 58 4.9. Water use efficiency of P. sidoides, on a dry root mass basis, in response to different soil water levels. 59 5.1. TLC analysis showing the presences of scopoletin (scop) and esculin (esc) in P. sidoides plants exposed to different water and nitrogen treatments. N = nitrogen in kg · ha-1, ADL = allowable depletion level of plant available water. 74 5.2. Principal component analysis (PCA) results of P. sidoides on NMR spectra, with the ellipse representing Hotelling within 95% confidence. The well watered treatments is represented by green colour, moderately stressed by blue colour and severely stressed by red colour. 75 5.3. Orthogonal partial least square discriminatory analysis (OPLS -DA) results of P. sidoides on NMR spectra, with the ellipse representing Hotelling within 95% confidence. The well watered treatments is represented by green colour, moderately stressed by blue colour and severely stressed by red colour. 76

viii

5.4. Hierarchical clustering showing the two separated groups from results OPLS-DA of P. sidoides plants exposed to different water treatments. The green group represents the well watered treatments with none of these treatments in the blue group (water stressed). 76 5.5. Score contribution plot showing NMR regions which had an influence on the separation of the well watered treatments (A) and the water stressed treatments (B), in the OPLS-DA. 78 5.6. Principal component analysis (PCA) results of P. sidoides on NMR spectra, with the ellipse representing Hotelling within 95% confidence. The Nitrogen treatments are represented as follows: no nitrogen (green), 50 (blue), 100 (brown) and 150 (yellow) kg · N · ha-1. 80 5.7. Orthogonal partial least square discriminatory analysis (OPLS -DA) results of P. sidoides on NMR spectra, with the ellipse representing Hotelling within 95% confidence. The Nitrogen treatments are represented as follows: no nitrogen (green), 50 (blue), 100 (brown) and 150 (yellow) kg · N · ha-1. 80

APPENDICES PAGE

1. Randomisation of the trial, with the blocks representing replicates. N = nitrogen in kg · ha-1 and allowable depletion level (ADL) as percentage (%) of plant available water. 91 2. The calibration functions developed to determine the soil water loss at 0 - 0.2 (A), 0.2 - 0.4 (B), 0.4 - 0.6 (C), 0.6 - 0.8 (D) and 0.8 - 1.0 m (E) of the soil profile. 93 3. Plant identification dispatch list: SANBI (South African National Biodiversity Institute). 95 4. Preparation of treatment plots under rainshelter (4.a and b), the trial after planting (4.c) and P. sidoides plants under well watered (4.d), moderately stressed (4.e) and severely stress conditions (4.f). 96

GROWTH, YIELD AND CHEMICAL COMPOSITION OF PELARGONIUM SIDOIDES DC. IN RESPONSE TO NITROGEN AND SOIL WATER MANAGEMENT

M.M. Mofokeng

ABSTRACT

Pelargonium sidoides DC. is one of many medicinal plant species that are harvested in the wild. The current trade in these medicinal plants has negatively affected their distribution in the wild due to unsustainable harvesting. The demand for medicinal plants is unlikely to decrease, but sustainability of the supply is questionable. This is because commercial exploitation threatens to deplete their populations, resulting in many species being considered vulnerable to extinction and being lost from their natural habitat. Increased demand, which is already too high to be met by sustainable harvesting, and price increases presents potential opportunities for cultivating indigenous medicinal plants at a commercial scale. A need for basic information on how to grow the plants and other related matters necessitates that field trials be conducted, before farmers could be expected to venture into cultivating medicinal plants and this study aimed at developing such information.

The specific objectives of the study were to investigate 1) the physiological and morphological, 2) the yield components and 3) chemical composition of P. sidoides in response to soil water and nitrogen levels. The study was conducted under a rainshelter as a randomized complete block design with three irrigation levels, four nitrogen levels and three replicates. The irrigation treatments were 30% allowable depletion level (ADL) (well watered treatment), 50% ADL (moderately stressed treatment) and 70% ADL (severely stressed treatment), while the nitrogen levels were 0, 50, 100 and 150 kg · N · ha-1. Dried root samples were analyzed for the presence of two standard compounds, scopoletin and esculin, using thin layer chromatography (TLC) and for metabolite profiling using the nuclear magnetic resonance technique (NMR).

Nitrogen and water level had no significant interaction effect on all measured parameters. Water stress significantly reduced stomatal conductance, while nitrogen had no significant effect on it. The well watered control had a significantly higher leaf area index, plant height and leaf area compared to the water stressed treatments. Nitrogen had a significant effect on the number of leaves, where 100 kg · N · ha-1 had a significantly higher number of leaves

compared to other nitrogen treatments. The well watered treatment again had a significantly higher total biomass, fresh and dry root yield; and nitrogen use efficiency compared to the water stressed treatments. The water use efficiency was significantly decreased in the well watered treatment. Water stress significantly increased nitrogen content and chlorophyll content of P. sidoides plants and leaves, respectively. Nitrogen levels of 50 and 100 kg · N · ha-1 resulted in a significantly higher total biomass compared to the control. TLC analysis showed the presence of the two standard compounds in all treatment samples analyzed. The orthogonal partial least square discriminatory analysis (OPLS-DA), which was performed on the NMR spectral data, showed separation between the irrigation treatments, resulting in two clusters representing the well watered treatment and the water stressed treatments. Asparagine, arginine, sucrose, xylose, glucose and citric acid were found to be the compounds associated with the separation. There was no separation of the samples regarding the nitrogen treatments which is indicative of the small effect of nitrogen on the metabolite content of the treatments. The results from this study showed a relationship between physiological, morphological and yield response as well as chemical composition of P. sidoides. The observed stomatal closure under water stress conditions, due to low turgor pressure in the guard cells, had a significant negative effect on leaf area, leaf area index, total biomass and root yield. Also the increases in total biomass and root yield under well watered conditions could be attributed to the increased primary metabolite content, under such conditions. The recommendation from the study is that P. sidoides plants should be grown under well watered conditions for a year or two to increase growth and root yield. Thereafter plants can be exposed to water stress in the second season, which is expected to increase the concentration of important secondary metabolites.

xii

CHAPTER 1

GENERAL INTRODUCTION

1.1. Background Africa, and especially southern Africa, is rich in plant diversity with recent statistics showing that about 25% of the total number of higher plants in the world is found in Africa, south of the Sahara (Van Wyk, 2008). With over 19 500 indigenous plant species in about 350 plant families, South Africa has the richest temperate flora in the world (Crouch et al., 2008).

The herbal medicine industry was, in 2004, already worth more than US$ 60 billion and could be a natural answer to some human ailments (Magoro, 2008). It is for this reason that herbal medicines are growing in popularity in wealthy countries and their use remain widespread in developing countries (WHO, 2004).

1.2. African medicinal plant industry Africa is well known as a rich source of medicinal plants and traditional rural African communities have relied upon the spiritual and practical skills of Traditional Health Practitioners for many years (Mativandlela, 2005). The traditional medicine trade in South Africa is a large and growing industry with about 27 million consumers and a contribution to the national economy was estimated at R2.9 billion already in 2007 (Sobiecki, 2014). Lewu et al. (2007) reported that 700 000 tonnes of plant materials were being harvested from the wild and consumed annually, with an estimated value of as much as 150 million US dollars per annum. Further reports are that at least 133 000 income earning opportunities were generated by the trade in traditional medicinal plants and their products in South Africa (Mander et al., 2007). There is a growing interest in natural and traditional medicines and scientists are looking for new cures in collaboration with traditional health practitioners (Mativandlela, 2005).

The current trade in medicinal plants in Africa has negatively affected the demography of most plant species harvested from the wild due to unsustainable harvesting (Lewu et al., 2006). The removal of whole plants, roots and bulbs totals 57% of all plant material traded in the Witwatersrand and may signify total loss of individual plants from the population (Dzerefos and Witkowski, 2001). Mander et al. (2007) reported that 86% of the plant parts harvested in the wild will result in the death of the entire plant and this has significant implications for the 1

sustainability of supply. Ring barking and uprooting are the most common methods of collection used by commercial gatherers, which kills the plants. This is because the active ingredients are usually contained in the bark and roots of plants (Cunningham, 1993; Taylor et al., 1996). On the other hand, traditional harvesting methods that were more conservative and usually protected the plants to some extent, are now being ignored as the pressure to generate income exceeds concern for the resource base (Taylor et al., 1996).

O’Connor (2004) reported that the demand for medicinal plants is unlikely to decrease, but it is questionable whether continued harvesting can be sustained. This is because commercial exploitation threatens to deplete medicinal plant populations, resulting in many species being considered vulnerable to extinction and being lost from their natural habitat (Mativandlela, 2005). Finding a balance between natural resource use and conservation of these plants can become a challenge when dealing with the commercial use of wild resources (Motjotji, 2011). Canter et al. (2005) further indicated that approximately two thirds of the 50 000 different medicinal plant species in use are collected from the wild, with only 10% being cultivated. Species differ in their vulnerability to harvesting because of differences in demand, the manner of harvesting, the abundance of the species in the absence of harvesting and in their life history attributes (O’ Connor, 2004). Vulnerability increases with low seedling survival, slow growth rates or long delays before harvestable size or reproductive maturity is reached (Dzerefos and Witkowski, 2001).

1.3. Rationale for the study The importance of cultivation as a way of conserving medicinal plants cannot be ignored. Diminishes in supply and price increases presents potential opportunities for cultivating indigenous medicinal plants (Mander, 1998). The demand for medicinal plants is already too high to be met by sustainable harvesting and the only real solution would be to develop medicinal plants as crops through small-scale farming (Sparg et al., 2005). A need for basic information on how to grow the plants, planting to harvesting period and yield; and other related matters necessitates that field trials be conducted to establish this information, before farmers could be expected to venture into cultivating medicinal plants (Sparg et al., 2005).

The commercialisation of P. sidoides has led to indiscriminate harvesting of wild populations to meet the growing demand, particularly for the export market while another threat to wild

populations is habitat loss due to the expansion of human settlements and the associated anthropogenic activities (Moyo et al., 2012).

Pelargonium sidoides has been harvested from the wild for decades by rural communities who depend on it for health purposes and also as a source of income. The current Government has introduced some interventions to help in dealing with the over-exploitation of the species, but this means cutting the main source of income for rural communities and in most cases leads to illegal harvesting and arrests. Commercial cultivation of P. sidoides will help in conserving the wild species while ensuring income for the rural communities. It is therefore very important to research the agronomic practices for P. sidoides which will increase yields while maintaining the medicinal value of the plant.

1.3.1. Study purpose and objectives The increasing demand for P. sidoides, coupled with diminishing supplies from the wild, legal restrictions as well as the associated health and income benefits necessitate commercial cultivation of the species. It is therefore important that technical scientific information be developed on the cultivation of P. sidoides. Therefore the study aimed to develop such information.

It has been reported that soil water content, which has a direct influence on nutrient content, has an effect on chemical composition of P. sidoides and thus the focus of this study was on these two production factors.

The objectives of the study were as follows:

 To determine the effect of different nitrogen levels and soil water depletion regimes on growth, yield and chemical composition of P. sidoides

 To investigate the interaction effect of nitrogen and soil water depletion regimes on growth, yield and chemical composition of P. sidoides

 To start developing guidelines for nitrogen fertilization and irrigation management of P. sidoides

1.3.2. Hypotheses i. There is an optimum N level (e.g. 100 kg · N · ha-1) that will improve growth and increase yield of storage roots with no negative effect on medicinally active compounds of the plant.

ii. Improved root yield will lead to increases in total yield of the medicinally active compounds.

iii. Increasing nitrogen rates will significantly increase chlorophyll content and leaf area.

iv. Appropriate irrigation scheduling will improve yield of storage roots with no negative effect on active compounds, and thus increase the total yield of active compounds.

1.3.3. Structure of dissertation This dissertation is divided into six chapters, as follows; Chapter one (1) covers the general introduction and background of wild medicinal plants. The medicinal plant industry and cultivation of medicinal plants are discussed. The rationale for the study, purpose and objectives; and hypothesis are included in this chapter.

Chapter two (2) is the literature review, with specific details on the botany and distribution of P. sidoides, medicinal properties, harvesting and uses of P. sidoides; propagation and cultivation of the species. Chapters 3, 4 and 5 present research results with each chapter focusing on specific research objectives. The chapters have been written in the format of papers for publication and having used this format, there are some repetition although efforts have been made to keep this to a minimum. Chapter three is focused on the physiological and morphological response of P. sidoides to water and nitrogen level. Chapter four presents the yield response of P. sidoides to water stress and nitrogen. Chapter five provides an investigation into the chemical composition of P. sidoides in response to water stress and nitrogen level. Chapter six is the general discussion on the findings of the study with concluding comments and final recommendations. 4

References

Canter, P.H., Thomas, H. and Ernst, E. 2005. Bringing medicinal plants into cultivation: opportunities and challenges for biotechnology. Trends Biotechnol. 4: 180 – 185

Crouch, N.R., Douwes, E., Wolfson, M.M., Smith, G.F. and Edwards, T.J. 2008. South Africa’s bioprospecting, access and benefit-sharing legislation: current realities, future complications, and a proposed alternative. S. Afr. J. Sci. 104: 355 – 366

Cunningham, A.B. 1993. African medicinal plants: Setting priorities at the interface between conservation and primary health care. People and plants working paper 1. Unesco. Paris.

Dzerefos, C.M. and Witkowski, E.T.F. 2001. Density and potential utilization of medicinal grassland plants from Abe Bailey Nature Reserve, South Africa. Biodivers. Conserv. 10: 1875 – 1896

Lewu, F.B., Grierson, D.S. and Afolayan. A.J. 2006. Clonal propagation of Pelargonium sidoides: A threatened medicinal plant of South Africa. Afr. J. Biotechnol. 2: 123-125

Lewu, F.B., Adebola, P.O. and Afolayan. A.J. 2007. Commercial harvesting of Pelargonium sidoides in the Eastern Cape, South Africa: Striking a balance between resource conservation and rural livelihoods. J. Arid Environ. 70: 380 – 388

Magoro, M.D. 2008. Traditional health practices and the sustainability of extinction-prone traditional medicinal plants. Master’s thesis. University of South Africa. uir.unisa.ac.za Accessed on 28th August 2009

Mander, M., 1998. Marketing of indigenous medicinal plants of South Africa: a case study in Kwazulu-Natal. Food and Agricultural Organization of the United Nations, Rome. www.fao.org

Mander, M., Ntuli, L., Diederichs, N. and Mavundla, K. 2007. Economics of the traditional medicine trade in South Africa. South African Health Review 189 – 196 www.hst.org.za. Accessed on 28th August 2009

Mativandlela, S.P.N. 2005. Evaluation of isolates and identified phenolics from Pelargonium sidoides against Mycobacterium tuberculosis, other bacteria and fungi. MSc thesis. University of Pretoria, South Africa.

Motjotji, L. 2011. Towards sustainability of harvesting the medicinal plant Pelargonium sidoides DC. (GERANIACEAE). MSc dissertation. University of Witwatersrand, South Africa. wiredspace.wits.ac.za . Accessed on 25th September 2014

Moyo, M, Finnie, J.F. and Van Staden, J. 2012. Topolins in Pelargonium sidoides micropropagation: do the new brooms really sweep cleaner? Plant Cell Tiss. Org. 110: 319 – 327

O’Connor, T.G., 2004. Influence of land use on populations of medicinal plant Alepidea amatymbica in the southern Drakensberg. S. Afr. J. Bot. 70: 319 – 322

Sobiecki, J.F. 2014. The intersection of culture and science in South African traditional medicine. Indo-Pacific J. Phenomenol. 14: 1 – 11

Sparg, S.G., Jäger, A.K, Magwa M.L. and van Staden, J. 2005. Cultivating the medicinal plant Merwilla natalensis as a crop: A small-scale farming approach. Outlook Agric. 34: 116 – 120

Taylor, F., Mateke, S.M. and Butterworth, K.J. 1996. A holistic to approach the domestication and commercialization of non-timber forest product. In Domestication and commercialization of non-timber forest products in agroforestry systems, Non- wood forest products 9. Food and Agriculture Organization of the United Nations, Rome. www.fao.org. Accessed on 28th August 2009

Van Wyk, B.-E. 2008. A broad review of commercially important southern African medicinal plants. J. Ethnopharmacol. 119: 342 – 355

World health organization, 2004. Medicinal plants – guidelines to promote patient safety and plant conservation for the US$ 60 billion industry. www.who.int/mediacentre/news Accessed on 28th August 2009

CHAPTER 2

LITERATURE REVIEW

2.1. Introduction

A few medicinal plants that are cultivated for the international trade industry include Warburgia salutaris (pepperbark tree), Siphonochilus aethiopicus (African ginger), Agathosma sp. (buchu), Harpagophytum procumbens DC. (devil’s claw), Pelargonium sidoides DC. (rooi rabasam) and Xysmalobium undulatum (Uzara) amongst others (Mativandlela, 2005;Van Wyk, 2011). Many of these plants, though cultivated commercially, are also harvested in the wild, leading to over utilization of natural habitats (Mativandlela, 2005).

As supply of medicinal plants diminishes and prices increase, the potential opportunities for cultivating indigenous medicinal plants have increased substantially (Mander, 1998). With the increase in non-specialist gatherers and commercial traders, the demand for medicinal plants is already too large to be met by sustainable harvesting and the only real solution would be to develop medicinal plants as crops through cultivation (Sparg et al., 2005). Sparg et al. (2005) mentioned that the need for basic information on how to grow the plants, planting time, harvesting period and yield, amongst others, necessitates that field trials be conducted to establish this information, before farmers could be expected to venture into cultivating medicinal plants. Cultivation of medicinal plants provides the potential to grow plants under environmental or physiological conditions that may promote the production of desired bioactive secondary metabolites (White, 2006).

Harvesting of medicinal plants for trade is a vital component of rural livelihoods, especially for women who are the most vulnerable sector of rural society and who may have no other means of income generation (Motjotji, 2011).

2.2. Pelargonium sidoides

Of the estimated 3000 medicinal plant species that are regularly used in traditional medicine in South Africa, only about 38 indigenous species have been commercialized to some extent (Van Wyk, 2008), while several others are wild harvested and in many cases over exploited for multi

million Rand informal markets. Amongst these species which are unsustainably harvested and over exploited, is Pelargonium sidoides DC. which is in the Geraniaceae family (White, 2006; Lewu et al., 2007a and Van Wyk, 2008). Brendler and Van Wyk (2008) reported P. sidoides as one of several geophytic species of the genus that are important traditional medicines in South Africa.

The Red List of South African plants (2009, 2011), lists P. sidoides as a ‘declining’ species in the conservation status and this means that although the plant is not yet critically endangered, there is a dire need for implementation of conservation measures (Moyo et al., 2012).

2.2.1. Botany and distribution P. sidoides is a small rosette-like perennial herb with tuberous rhizomes, sparsely branched stems, rounded to heart shaped leaves and small tubular flowers that are dark maroon red to almost black (van Wyk, 2008 and Lewu et al., 2010). At full bloom, the species is about 20-50 cm high, forming a spiral of heart shaped leaves around its base (Lewu et al., 2010).

Morphologically it is very similar to its close relative Pelargonium reniforme, but can be readily distinguished by its dark red rather than pink petals (Lewu et al., 2007b). The sepals of P. sidoides are green with white margins, while those of P. reniforme are red with pink margins (White, 2006). The leaves of P. sidoides are velvety, cordate, heart-shaped and long-stalked, whereas those of P. reniforme are kidney shaped (Bladt and Wagner, 2007; Adewusi and Afolayan, 2009) and have crenate or finely lobed margins (Kolodziej, 2002). P. sidoides have yellow-green pollen whereas P. reniforme has whitish-green pollen (Bladt and Wagner, 2007). Although the shape of leaves and color of flowers can distinguish between the two species (Figure 2.1. A, B, D and E), the existence of gradual variations needs to be acknowledged (Kolodziej, 2002 and Lewu et al., 2007b).

A B C

D E F

Figure 2.1. Dark maroon flowers (A) of P. sidoides as compared to the pink flowers of P. reniforme (D). The leaves of P. sidoides (B) and those of P. reniforme (E). Tuberous roots of P. sidoides (C) with their internal reddish colour (F).

P. sidoides usually grows in short grassland and sometimes on stony soil, varying from sand to clay-loam, shale or basalt; and it is usually restricted to the Grassland and Savanna biomes and Grassy fynbos (Mativandlela, 2005 and Motjotji, 2011). The thickened underground root system (Figure 2.1. C) penetrates deep into the ground and seem to be a special adaptation which enables the plant to survive grass fires which occurs throughout much of its habitat range (Motjotji, 2011). The roots of P. sidoides start off pale pink and as the plant matures turn deep red (Figure 2.1. F) and it is only at this stage that it has any commercial value (Stern, 2008). It grows naturally in summer rainfall areas with average annual rainfall ranging from 200 to 800 mm per annum (Mativandlela, 2005).

The species occurs naturally in Lesotho as well as the Eastern Cape, Free State, Northwest, Gauteng and Mpumalanga Provinces of South Africa (Figure 2.2) (Brendler and van Wyk, 2008), thus it can tolerate a wide range of environmental conditions (Motjotji, 2011). P. reniforme has a much narrower distribution range and is more or less confined to the Eastern 10

Cape Province (Brendler and van Wyk, 2008). P. sidoides is used locally in the Eastern Cape for the treatment of several cold related ailments, in man and livestock. There has been a significant increase in demand for the plant for both local uses and international pharmaceutical producers in recent years, resulting in an increase in the number of collectors and rate of wild harvesting (Lewu et al., 2007a). The commercial demand for P. sidoides tubers for local and international trade has caused an enormous increase in the rate of uncontrolled, illegal and indiscriminate wild harvesting, which is fast leading to irreparable reductions in the number of natural populations (Colling et al., 2010). Two popular articles in newspapers and one online publication reported uncontrolled harvesting of at least 20 tonnes of P. reniforme and P. sidoides roots in the Eastern Cape in 2002 (White et al., 2008).

Figure 2.2. Geographical distribution of P. sidoides (maroon circles) and P. reniforme (pink squares) in South Africa. Source: SANBI distribution data (2013). 11

2.2.2. Bio-activity and uses of P. sidoides Various studies have been done on the activity of P. sidoides extracts against acute bronchitis and common colds (Golovatiouk and Chuchalin, 2002; Chuchalin et al., 2005; Lizogub et al.,2007; Agbabiaka et al., 2008 and Bachert et al., 2009); on the antibacterial, antifungal and antitubercular activity (Kolodziej et al., 2003; Seidel and Taylor, 2004; Mativandlela, 2005; Lewu et al, 2006; Conrad et al., 2007; Mativandlela et al., 2007; Wittschier et al., 2007 and Kim et al., 2009); antimicrobial, antiviral and immunomodulatory activity (Kayser et al., 2001; Kolodziej and Kiderlen, 2007; Schnitzler et al., 2008 and Kolodziej, 2011). Bladt and Wagner (2007), gives a good account of the history of P. sidoides in curing tuberculosis.

The fleshy bright red tubers have been widely used to treat diarrhoea and dysentery (Kolodziej and Kiderlen, 2007; Lewu et al., 2007a; Brendler and van Wyk, 2008); bronchitis (Chuchalin et al., 2005 and Van Wyk, 2011), cough, fever, tuberculosis, sore throat, fatigue and weakness of the body (Mativandlela, 2005; Lewu et al., 2006; White et al., 2008; Van Niekerk and Wynberg, 2012).

The tuberous roots are the raw material for an important German phytomedicine (Lewu et al, 2007a and Van Wyk, 2008) and many other herbal medicine products, which are approved and registered drugs in countries such as Turkey, Mexico, Brazil, Austria, Belgium, Italy and many others, for the treatment of common colds, acute bronchitis and infections of the upper respiratory tract (Chuchalin et al., 2005; Kolodziej, 2007; European medicines agency, 2012; Street and Prinsloo, 2013). In Cape Town, South Africa, a cough remedy called Linctagon (Figure 2.3) is manufactured for the local market (Stern, 2008). There is a possibility that the key ingredient may be revived as a cure for TB, particularly after the rise of drug-resistant strains of the disease (Stern, 2008). The annual sales of P. sidoides, one of the ten commercially important medicinal plants of South Africa, exceed € 80 million in Germany alone and most of the material used is still harvested from the wild by locals from rural communities in South Africa (Street and Prinsloo, 2013).

Figure 2.3. Some of the pharmaceutical products, made from root extracts of P. sidoides that are available over the counter in South Africa. Source: www.nativa.co.za

2.2.3. Wild harvest of P. sidoides Most of the gatherers in the Eastern Cape are from rural communities, with no other source of income and thus they rely heavily on natural resources for sustenance (Lewu et al., 2007a; Van Niekerk and Wynberg, 2012). In their findings, Lewu et al. (2007a) reported that more than 26 tonnes of P. sidoides were collected and an estimated $14 000 worth of the species sold from 10 settlements in the Eastern Cape, in just one month. Although harvesting and sales of P. sidoides from the wild brings financial returns to the people, the consequences of uncontrolled harvesting may affect its availability for future generations (Lewu et al., 2007a). All the respondents in the mentioned study (Lewu et al., 2007a) agreed that P. sidoides was becoming more difficult to find in the wild. The Eastern Cape Department of Economic Development and Environmental Affairs, in June 2006, responded to over-exploitation of pelargonium by placing a temporary ban on harvesting and exporting of P. sidoides and P. reniforme from the wild, and similar safeguards have also been placed by the Lesotho Government (African Centre for Biosafety, 2008). The pressure to supply the roots has prompted illegal harvesting, resulting in several illegal harvesters being arrested in the Eastern Cape and in Lesotho, already in 2008 (African Centre for Biosafety, 2008).

Lewu et al. (2007a) observed that the rate of harvesting does not support annual collection, as the second year of a regeneration study gave less than 50% biomass yield compared to the initial harvested mass and therefore recommended that the species be left undisturbed for some years after harvesting to give the population time to recover. Motjotji (2011) also reported that

P. sidoides roots may require 10 – 15 years of recovery to reach the same pre-harvest tuber mass, thus any harvesting before the suggested period has elapsed could have a negative effect on wild populations.

Harvesters collect large tubers with dark-red colouration as these tubers are believed to be more potent and thought to have higher concentrations of the desired medicinal compounds (Motjotji, 2011). However, White (2006) found no direct relationship between darkness of colour and umckalin concentration in dried roots of P. sidoides. Motjotji (2011) argues that the lack of any significant relationship between root colour and umckalin concentration may be due to the use of a single extract (umckalin) by White (2006), as opposed to a complex of extracts reported by the Pelargonium industry. It is suspected that umckalin may not be an active compound of P. sidoides but just a marker compound to differentiate between roots of P. sidoides and those of P. reniforme. Furthermore, although P. reniforme does not have umckalin, both the roots of P. sidoides and P. reniforme are traditionally used for the same purposes (Kolodziej, 2002; White, 2006 and White et al., 2008). Mativandlela (2005) found that the roots of P. reniforme showed inhibitory activity against M. tuberculosis. The same study justified the traditional use of the two Pelargonium species in the treatment of lung and respiratory tract infections. P. sidoides has significantly higher yield of coumarins than P. reniforme. The total coumarin content of P. sidoides roots is approximately 0.05% related to dry weight, with umckalin accounting for about 40% of total coumarin content (European medicines agency, 2012), which could be a possible reason why P. sidoides roots are preferred more than those of P. reniforme. The coumarin pattern of EPs®, a proprietary root extract contained in one of the herbal products Umckaloabo®, was strongly reminiscent of that of P. sidoides (Gödecke et al., 2005 and Kolodziej, 2007).

2.2.4. Propagation and cultivation Although P. sidoides tubers are highly sought-after, they are only cultivated on a small, virtually negligible scale (Colling et al., 2010). Some of the reasons for the small scale plantation include economies of scale and slow regeneration rate of the plant (Van Niekerk and Wynberg, 2012). To secure a regular supply of P. sidoides large scale cultivations were initiated by Schwabe, a German Pharmaceutical company, in Kenya and Mexico, when permitting arrangements were becoming stricter in southern Africa (Van Niekerk and

Wynberg, 2012). Van Wyk (2011) mentions suitability for cultivation as one of the important aspects when selecting species with commercial potential as herbal products.

Lewu et al. (2010) found that the germination of P. sidoides seeds is dependent on light, temperature, pre-chilling, seed age and source of seed; and this could be a possible explanation for the low rate of regeneration by seed in the wild (Lewu et al., 2006). They suggested that aerial vegetative parts, which are normally discarded during harvesting, can be used for propagation of this species (Lewu et al., 2006).

Naturally, regeneration of the species is by seed and from development of perennates through an underground root system (Lewu et al., 2006). However, viability of the seeds collected from the wild is low, coupled with very low seed germination and during harvest it is the roots that are removed, threatening the opportunity for perennial rejuvenation (Lewu et al., 2006). Canter et al. (2005) also mentioned that cultivation of some herbal plants has proved to be difficult because of low germination rates, specific ecological requirements or lack of knowledge. The increased rate of harvesting and number of plant gatherers has made its exploitation unsustainable (Lewu et al., 2006).

Lewu et al. (2006) reported that cultivation of P. sidoides could be an alternative source of supply in order to take pressure off the wild stock. The same sentiment is shared by White et al. (2008), mentioning that commercial cultivation of pelargonium has distinct advantages which would be ensuring a supply of roots of the correct species that have documented medicinal efficacy, and that the use of irrigation can increase the biomass yield and decrease chemical variability of plants. One of the practical measures against over-exploitation of plant natural resources is to improve propagation and cultivation (Lewu et al., 2006). In the best interest of conserving P. sidoides and the sustainable supply of roots for local and international markets, White et al. (2008) strongly recommended that cultivation be further researched and pursued as a promising alternative to wild harvest.

However, the challenge with propagation and cultivation of P. sidoides is that plants propagated from the wild stock revealed that the concentration of umckalin in their roots was significantly lower than that of wild plants, while the concentration of cultivated non-clonal plants compared to that of wild plants (White, 2006). In analyzing root extracts from different

regions, White (2006) revealed that average root umckalin concentrations were strongly related to both the average soil pH and annual rainfall of the collection sites. The study further found that prolonged water stress treatments did not significantly alter the root umckalin concentrations of cultivated plants relative to well-watered controls. Furthermore, the umckalin concentration in new roots of cultivated plants was not significantly different to that of old roots collected in the wild (White, 2006).

References

Adewusi, E.A. and Afolayan, A.J. 2009. Safety evaluation of the extract from the roots of Pelargonium reniforme Curtis in male wistar rats. Afr. J. Pharm Pharmacol. 8: 368 – 373

African Centre for Biosafety, 2008. Knowledge not for sale: Umckaloabo and the Pelargonium patent challenges. www.biosafetyafrica.org.za. Accessed on 10th October 2009

Agbabiaka, T.B, Guo, R. and Ernst.E. 2008. Pelargonium sidoides for acute bronchitis: A systematic review and meta-analysis. Phytomedicine 15: 378 – 385

Bachert, C., Schapowal, A., Funk, P. and Kieser, M. 2009. Treatment of acute rhinosinusitis with the preparation from Pelargonium sidoides EPs 7630: a randomized, double- blind, placebo-controlled trial. Rhinology 47: 51 – 58

Bladt, S. and Wagner, H. 2007. From the Zulu medicine to the European phytomedicine Umckaloabo®. Phytomedicine 14: 2 – 4

Brendler, T. and van Wyk, B.-E. 2008. A historical, scientific and commercial perspective on the use of Pelargonium sidoides (Geraniaceae). J. Ethnopharmacol. 119: 420 – 433

Canter, P.H., Thomas, H. and Ernst, E. 2005. Bringing medicinal plants into cultivation: opportunities and challenges for biotechnology. Trends Biotechnol. 4: 180 – 185

Chuchalin, A.G., Berman, B., and Lehmacher, W. 2005. Treatment of acute bronchitis in adults with a Pelargonium sidoides preparation (EPs® 7630): a randomized, double- blind, placebo-controlled trial. Explore 6: 437 – 445

Colling, J., Groenewald, J.H. and Makunga, N.P. 2010. Genetic alterations for increased coumarin production lead to metabolic changes in the medicinally important Pelargonium sidoides DC. (Geraniaceae). Metab. Eng. 12: 561 – 572

Conrad, A., Jung, I., Tioua, D., Lallemand, C., Carrapatoso, F., Engels, I., Daschner, F.D. and Frank, U. 2007. Extract of Pelargonium sidoides (EPs® 7630) inhibits the interactions of group A-streptococci and host epithelia in vitro. Phytomedicine 14: 52 – 59

European medicines agency, 2012. Assessment report on Pelargonium sidoides DC. and/or Pelargonium reniforme Curt., radix. EMA/HMPC/560962/2010. www.ema.europa.eu Accessed on 10th October 2009

Gödecke, T., Kaloga, M. and Kolodziej, H. 2005. A phenol glucoside, uncommon coumarins and flavonoids from Pelargonium sidoides DC. Z. Naturforsch. 60b: 677 – 682

Golovatiouk, A. and Chuchalin, A.G. 2002. Efficacy of an extract from Pelargonium sidoides (EPs 7630) compared to placebo in patients with acute bronchitis. Phytopharmaka 7: 3 – 12

Kayser, O., Kolodziej, H. and Kiderlen, A.F. 2001. Immunomodulatory principles of Pelargonium sidoides. Phytother. Res. 15: 122 – 126

Kim, C.E., Griffiths, W.J. and Taylor, P.W. 2009. Components derived from Pelargonium stimulate macrophage killing of Mycobacterium species. J. Appl. Microbiol. 106: 1184 – 1193

Kolodziej, H. 2002 Pelargonium reniforme and Pelargonium sidoides: their botany, chemistry and medicinal use. In: Lis-Balchin, M. (ed), Geranium and Pelargonium: The genera Geranium and Pelargonium. Taylor and Francis, London, pp 262 - 290

Kolodziej, H, Kayser, O, Radtke, O.A., Kiderlen, A.F. and Koch, E. 2003. Pharmacological profile of extracts of Pelargonium sidoides and their constituents. Phytomedicine 10: 18 – 24

Kolodziej, H. and Kiderlen, A.F. 2007. In vitro evaluation of antibacterial and immunomodulatory activities of Pelargonium reniforme, Pelargonium sidoides and the related herbal drug preparation EPs® 7630. Phytomedicine 14: 18 – 26

Kolodziej, H. 2007. Fascinating metabolic pools of Pelargonium sidoides and Pelargonium reniforme, traditional and phytomedicinal sources of the herbal medicine Umckaloabo®. Phytomedicine 14: 9 – 17

Kolodziej, H. 2011. Antimicrobial, antiviral and immunomodulatory activity studies of Pelargonium sidoides (Eps® 7630) in the context of health promotion. Pharmaceuticals 4: 1295 – 1314

Lewu, F.B., Grierson, D.S. and Afolayan. A.J. 2006. Clonal propagation of Pelargonium sidoides: A threatened medicinal plant of South Africa. Afr. J. Biotechnol. 2: 123-125

Lewu, F.B., Adebola, P.O. and Afolayan, A.J. 2007a. Commercial harvesting of Pelargonium sidoides in the Eastern Cape, South Africa: Striking a balance between resource conservation and rural livelihoods. J. Arid Environ. 70: 380 – 388

Lewu, FB., Grierson, D.S. and Afolayan, A.J. 2007b. Morphological diversity among accessions of Pelargonium sidoides DC. in the Eastern Cape, South Africa. Genet. Resour. Crop Evol. 54: 1 – 6

Lewu, FB., Grierson, D.S. and Afolayan, A.J. 2010. Influences of seed source, pre-chilling, light and temperature on the germination of South African Pelargonium sidoides. J. Agric. Sci. Technol. 3: 18 – 23

Lizogub, V.G, Riley, D.S., and Heger, M. 2007. Efficacy of a Pelargonium sidoides preparation in patients with the common cold: a randomized, double blind, placebo- controlled clinical trial. Explore 3: 573 – 584

Mander, M., 1998. Marketing of indigenous medicinal plants of South Africa: a case study in Kwazulu-Natal. Food and Agricultural Organization of the United Nations, Rome. www.fao.org. Accessed on 10th October 2009

Mativandlela, S.P.N., Meyer, J.J.M., Hussein, A.A. and Lall, N. 2007. Antitubercular activity of compounds isolated from Pelargonium sidoides. Pharm. Biol. 45: 645 – 650

Moyo, M, Finnie, J.F. and Van Staden, J. 2012. Topolins in Pelargonium sidoides micropropagation: do the new brooms really sweep cleaner? Plant Cell Tiss. Org. 110: 319 – 327

Schnitzler, P, Schneidera, S., Stintzingb, F.C., Carleb, R. and Reichlingc, J. 2008. Efficacy of an aqueous Pelargonium sidoides extract against herpesvirus. Phytomedicine 15: 1108 – 1116

Seidel, V. and Taylor, P.W. 2004. In vitro activity of extracts and constituents of Pelagonium against rapidly growing mycobacteria. Int. J. Antimicrob. Ag. 23: 613 – 619

Sparg, S.G., Jäger, A.K, Magwa M.L. and van Staden, J. 2005. Cultivating the medicinal plant Merwilla natalensis as a crop: A small-scale farming approach. Outlook Agric. 34: 116 – 120

Stern, J. 2008. Big medicine from a little plant. MediaClubSouthAfrcia.com. Accessed in September 2013

Street, R.A. and Prinsloo, G. 2013. Commercially important medicinal plants of South Africa: A review. J. Chem. Article ID 205048

Van Niekerk, J. and Wynberg, R. 2012. The trade in Pelargonium sidoides: Rural livelihood relief or bounty for the ‘bio-buccaneers’? Dev. South. Afr. 29: 530 – 547

Van Wyk, B.-E. 2008. A broad review of commercially important southern African medicinal plants. J. Ethnopharmacol. 119: 342 – 355

Van Wyk, B.-E. 2011. The potential of South African plants in the development of new medicinal products. S. Afr. J. Bot. 77: 812 – 829

Wittschier, N., Faller, G. and Hensel, H. 2007. An extract of Pelargonium sidoides (EPs 7630) inhibits in situ adhesion of Helicobacter pylori to human stomach. Phytomedicine 14: 285–288

White, A.G. 2006. The effect of geography, cultivation and harvest technique on the umckalin concentration and growth of Pelargonium sidoides (Geraniaceae). MSc thesis. Rhodes University, South Africa.

White, A.G., Davies-Coleman, M.T. and Ripley, B.S. 2008. Measuring and optimising umckalin concentration in wild-harvested and cultivated Pelargonium sidoides (Geraniaceae). S. Afr. J. Bot. 74: 260–267

CHAPTER 3

PHYSIO-MORPHOLOGICAL RESPONSE OF PELARGONIUM SIDOIDES DC. TO SOIL WATER AND NITROGEN LEVEL

Abstract

Water stress is the most limiting factor in agricultural productivity in arid and semi-arid regions and causes very high losses in crop yield. Regulation of growth and stomatal conductance are the main mechanisms by which plants respond to water stress. Pelargonium sidoides is a medicinal plant that grows in South Africa and is used for the treatment of upper respiratory ailments. Cultivation has been considered as a viable means of reducing the pressure on natural populations of this species, but little to or no information is available in this regard. Water and nitrogen supply are two of the most important factors that affect growth and yield of plants. This study therefore aimed at investigating the physiological and morphological response, in relation to growth, of P. sidoides to soil water and nitrogen levels. To achieve this objective P. sidoides plants were grown under a rainshelter and exposed to three irrigation levels (well watered control, moderate water stress and severe water stress treatment) and four nitrogen levels (0, 50, 100 and 150 kg · N · ha-1). Nitrogen and water level had no significant interaction effect on all measured parameters. Water stress significantly reduced stomatal conductance, while nitrogen had no significant effect on it. The well watered control had a significantly higher leaf area index, plant height and leaf area compared to the water stressed treatments. Nitrogen had a significant effect on the number of leaves, where 100 kg · N · ha-1 had a significantly higher number of leaves compared to other nitrogen treatments. The study provides a first report on the response of P. sidoides to water and nitrogen; and showed that the plant responds to water stress by closing its stomata and employing other morphological strategies like reducing plant growth. These results will assist in explaining yields under different irrigation and nitrogen management strategies in cultivation of P. sidoides.

Keywords: allowable depletion, growth response, nitrogen, Pelargonium sidoides, stomata, water stress This chapter has been published in the South African Journal of Botany: M.M. Mofokeng, J.M. Steyn, C.P. du Plooy, G. Prinsloo, H.T. Araya. 2015. Growth of Pelargonium sidoides DC. in response to water and nitrogen level. S. Afr. J. Bot. 100: 183-189

3.1. Introduction

Pelargonium sidoides, which grows naturally in South Africa, is used in the Eastern Cape Province for the treatment of several cold related ailments in humans and livestock (Lewu et al., 2006; Brendler and van Wyk, 2008). Furthermore, Helfer et al. (2014) also proposed that P. sidoides root extract has shown anti-HIV-1 properties. Up to now P. sidoides plant material for medicinal use has almost exclusively been harvested from the wild. However, there has been an increase in demand for the plant for traditional use as well as by local and international pharmaceutical companies (Lewu et al., 2007). As a result, cultivation has been considered as a viable means of reducing the pressure on natural P. sidoides populations. Information on the cultivation of medicinal plants such as P. sidoides is, however, very limited and therefore further research was needed.

Water and nutrient supply are two of the most important factors that can affect growth, biomass yield and chemical composition of plants and therefore this study focused on these two production factors. Exchange of water and carbon dioxide (CO2) between leaves and the ambient air are important plant processes by which heat is dissipated through transpiration, while a primary substrate for photosynthesis is taken up (Streck, 2003). The ability of plants to adjust gaseous exchange through stomata permits them to control water relations and carbon assimilation; and the opening of the stomatal pore reflects a compromise between the photosynthetic requirement for CO2 and the availability of water (Tricker et al., 2005). Regulation of leaf expansion and stomatal conductance are the main mechanisms by which plants respond to soil water deficit (Liu and Stützel, 2002; Eiasu et al., 2012). High transpiration causes stomatal closure, possibly by increasing the water potential gradient between the guard cells and other epidermal cells or by lowering the leaf water potential, either of which directly decrease the turgor pressure of guard cells relative to other epidermal cells or affect hormonal distribution (Bunce, 1996).

Water deficit in plants leads to physiological disorders, such as a reduction in photosynthesis and transpiration (Petropoulos et al., 2008); however the effects vary between species (Karkanis et al. 2011). Both photosynthesis and transpiration, which are closely related to dry matter production, are regulated by a stomatal feedback control mechanism which, in turn, is influenced by water deficits (Kumar et al., 1994 and Bota et al., 2004). The limitation of plant growth enforced by low water availability is mainly due to decreases in plant carbon balance,

which is dependent on the balance between photosynthesis and respiration (Flexas et al., 2006). An early response to water stress supports immediate survival, whereas acclimation, calling on new metabolic and structural capabilities mediated by altered gene expression, help to improve plant functioning under stress (Chaves et al., 2002). Shoot growth is more sensitive to water deficit than root growth and the mechanisms underlying the sustained root growth under water stress include osmotic adjustment and an increase in the loosening capacity of the cell wall (Chaves et al., 2002).

Deficit irrigation is becoming an important strategy to reduce agricultural water use in arid and semi-arid regions (Ayana, 2011). It is the practice of deliberately under irrigating crops to reduce water consumption while minimizing adverse effects of extreme water stress on yield (Ayana, 2011). Deficit irrigation does not always decrease yield, as deficits properly applied in some development stages may even increase crop yield (Bilibio et al., 2011).

Nitrogen increases leaf area index (LAI) and also improves the physiological properties of the plant (Kara and Mujdeci, 2010). Nitrogen is a component of many biological compounds that plays a major role in photosynthetic activity and crop yield capacity; and its deficiency constitutes one of the major yield limiting factors for production (Hokmalipour and Darbandi, 2011). Nitrogen deficiency leads to loss of green colour in the leaves, decreased leaf area and intensity of photosynthesis, leading to reduced photosynthate production and thus lower yields (Alva, et al., 2006, Bojović and Marković, 2009). Leaf area influences the interception and utilization of solar radiation of crop canopies (Hokmalipour and Darbandi, 2011), but it also plays an important role in water use (Liu and Stützel, 2002). Leaf area index is a significant feature for the determination of plant photosynthetic activity and is a crucial structural characteristic of plants due to the role of green leaves in controlling many biological and physical processes in plant canopies (Kara and Mujdeci, 2010). Over application of nitrogen causes many environmental pollution problems (Lee et al., 2011) and can lead to decreased yields due to luxury consumption (Alva et al., 2006).

Plants take up inorganic nitrogen contained in the water absorbed from soil solution through their root systems and thus, the fate of nitrogen is certainly coupled to that of water reaching the soil in the root zone (Alva et al., 2006). Water and nitrogen deficiency induces alterations of many morphological and physiological processes (Shangguan et al., 2000). Information on

the response of P. sidoides to different water stress and nitrogen deficiency levels is not known and thus the objective of this study was to investigate the effect of water stress and nitrogen level on the physiology and morphology of P. sidoides.

3.2. Materials and methods 3.2.1. Study area A field trial was conducted under a rainshelter at the Agricultural Research Council-Roodeplaat Vegetable and Ornamental Plant Institute (ARC-Roodeplaat VOPI), Pretoria, South Africa (25° 59’S ; 28° 35’E and 1 200 m.a.s.l.) during 2012 and 2013. The rainshelter is designed to automatically open when there is no rain and close during a rainfall event, thus excluding rainfall from the experiment. The physical and chemical properties of the soil at the experimental site are shown in Tables 4.1 and 4.2, respectively, while a summary of the weather data recorded by a weather station (Campbell Scientific, USA) at the experimental site during the experiment period is shown in Table 3.3.

Table 3.1. Physical properties of the soil at the experimental site, per 20 cm depth increment to a depth of 100 cm. Soil depth Sand Silt Clay Texture PWP FC BD (cm) % (mg/m3)

0 - 20 78 6 16 sandy loam 10.3 19.9 1.59 20 - 40 70 8 22 sandy clay loam 12.9 25.5 1.56 40 - 60 64 8 28 sandy clay loam 15.2 25.3 1.45 60 - 80 62 8 30 sandy clay loam 14.7 28.0 1.38 80 - 100 60 8 32 sandy clay loam 15.1 28.1 1.35 *PWP – permanent wilting point, FC – field capacity, BD – bulk density

Table 3.2. Chemical properties of the soil at the experimental site.

Soil depth (cm) Fe Mn Cu Zn Ca Mg Na P K Total N pH H2O mg/kg % Top soil 13.74 44.10 9.24 14.00 980 298 24.7 80.9 134 0.028 7.26 (0–20) Sub soil 9.74 28.50 5.64 7.43 1201 370 39.4 60.4 94 0.026 7.44 (20–40)

Table 3.3. Summary of weather data collected during the experiment period.

Month Temperature (°C) Wind Relative humidity VPD Rainfall ET0 (2013) (kPa) speed (%) (ms-1) mm Average Total max min max min Jan 30.73 16.85 0.90 87.45 34.61 1.28 90.3 168.0 Feb 32.06 15.77 0.81 88.80 27.89 1.40 35.0 159.2 Mar 29.53 14.43 0.75 89.78 30.54 1.11 75.9 133.2 Apr 26.10 9.21 0.67 91.70 32.39 0.89 98.2 107.6 May 24.83 4.97 0.59 89.73 23.43 0.87 0.55 95.3 *Max: maximum, min: minimum, VPD: vapor pressure deficit.

3.2.2. Trial design, plant material and duration The trial was a randomized complete block design with three replicates and 12 treatment plots per replicate. The treatment factors were water and nitrogen levels. Each treatment plot was 4.5 m2 in size, with 30 plants planted at a spacing of 0.5 m between the rows and 0.3 m in the row, giving a total plant population of 66 666 plants · ha-1.

The mother material was acquired from a nursery at the Golden Gate Highlands National Park, in the Free State province of South Africa, in 2010 and grown under shade-net (grey, 40% shade effect) at ARC-Roodeplaat VOPI. Root cuttings were made from the mother plants between January and February 2012. Four months old rooted cuttings of P. sidoides were transplanted into the rainshelter in May 2012 and were established for six months, to ensure a good stand, before application of the treatments. The trial was harvested in June 2013.

3.2.3. Treatment application Irrigation treatments were scheduled based on the allowable depletion levels (ADL) of the plant available water (PAW) in the soil. From previous observation the effective rooting depth (ERD) was determined as 400 mm. The predetermined treatments applied were 30% (well watered treatment), 50% (moderately stressed treatment) and 70% ADL (severely stressed treatment) of PAW, where a specific percentage of PAW was allowed to deplete from the ERD before filling the soil profile back to field capacity. A neutron probe (Waterman, Probe Version 1.6, 2005, Geotech) was used to monitor soil water content twice a week. Neutron probe readings were taken at intervals of 0.2 m to a soil depth of 1.0 m, through aluminum access 26

tubes that were pre-installed in each plot. The instrument was calibrated to a depth of 1.0 m, at intervals of 0.2 m and calibration functions were developed following the method described by Evett (2008). These calibration functions are given in Appendix 2.

The allowable soil water deficits for the three water levels were calculated as follows and they were based on the field capacity (FC), permanent wilting point (PWP) and bulk density (BD) as shown on Table 3.1.

PAW (0 – 200 mm) = ((FC – PWP) x BD) x 200 mm = ((0.199 – 0.103) x 1.59) x 200 mm = 30.5 mm

PAW (200 – 400 mm) = ((FC – PWP) x BD) x 200 mm = ((0.255 – 0.129) x 1.56) x 200 mm = 39.3 mm

Total PAW for ERD (400 mm) = 30.5 mm + 39.3 mm = 69.8 mm 30% ADL of PAW = 69.8 mm x 30% = 20.94 mm 50% ADL of PAW = 69.8 mm x 50% = 34.9 mm 70% ADL of PAW = 69.8 mm x 70% = 48.86 mm

A computerized, non-pressure regulated, drip irrigation system (NETAFIM, South Africa) with a discharge rate of 2 L per hour and maximum pressure of 270 kPa was used for irrigation. Soil samples were taken in the field to determine the soil properties and the nutrient status of the soil which are given in Tables 4.1 and 4.2 respectively.

The nitrogen (N) treatments were different levels of N, applied at the following rates; control (0), 50, 100 and 150 kg · N · ha-1. The N source used was Limestone ammonium nitrate (LAN 28% N), applied in two split applications of 50% each. The first N application was four months after planting, followed by the second application eight weeks after the first application.

3.2.4. Agronomic practices As a base application of potassium (K) and phosphorus (P) were applied and worked into the soil five days after planting, while nitrogen (N) was applied six weeks after planting, based on soil nutrient status and estimated nutrient requirements of rose-scented geranium (Pelargonium capitatum) (Araya et al., 2006), since there was no recommendation available for P. sidoides. Potassium (K) was applied as potassium chloride (50% K) at a rate of 110 kg · K · ha-1 and P was applied as single-super phosphate (11% P) at a rate of 30 kg · P · ha-1. Before nitrogen treatments were initiated, two weeks after planting, all plots were fertigated with potassium nitrate (13 % N, 38 % K) at a rate of 150 kg · N · ha-1 as a once off to boost plant establishment.

Weed control was performed manually with hand hoes. Servus (active ingredient: deltamethrin (pyrethroid)) was applied for pest control only when pests were identified after scouting.

3.2.5. Data collection and statistical analysis Leaf area index (LAI) was measured non-destructively, using a LAI 2200 plant canopy analyzer (Li-Cor Bioscience, USA). The instrument uses measurements made above and below the canopy to calculate light interception at five zenith angles, from which LAI is computed using a model of radiative transfer in vegetative canopies. One above canopy reading and four below canopy readings were taken using the 270° view cap; and this was replicated two times in each plot. Plant height (cm) was measured and the number of leaves were counted manually. LAI and plant height measurements were taken on a monthly basis after treatment implementation. After harvesting, the total leaf area per plant was measured with a leaf area meter (Li-3100, Li-Cor Inc., Lincoln, USA).

Stomatal conductance, which is a measure of the rate of passage of carbon dioxide (CO2) or water vapour through the stomata of a leaf, was measured using the SC-1 Leaf porometer (Decagon Devices, USA), on a monthly basis. Stomatal conductance is described as a function of the density, size, and degree of opening of stomata. The measurements were taken on the abaxial (bottom) side of a fully expanded mature leaf, during midday when the environmental factors were at their peak.

Data was subjected to analysis of variance (ANOVA) using GenStat® version 11.1 (Payne et al., 2008). Treatment means were separated using Fisher’s protected T-test for least significant differences (LSD) at 5% level of significance (Snedecor and Cochran, 1980).

3.2.6. Scanning Electron Microscope (SEM) observations Three leaf samples were collected and observed under an electron microscope to observe the stomata, following the method described by Motsa (2006) and Eiasu et al. (2012). The samples comprised of young, mature and old leaves. Samples (10 mm x 10 mm) were cut from each leaf sample and fixed in glutaraldehyde (3% w/v) immediately after cutting it from the plant. They were then rinsed thoroughly with a phosphate buffer (0.1 M, pH 7.0) for 15 minutes, and repeated three times. Thereafter the samples were dehydrated in ethanol series (30 – 100% w/v) and then dried in a critical point drying apparatus (Bio-Rad E300, Watford, England). The dried samples were mounted on copper stubs and coated with gold in a vacuum coating unit (Polaron E5200C, Watford, England). The samples were then observed under a JSM 840 scanning electron microscope (JEOL, Tokyo, Japan) at 2000X magnification.

3.3. Results and discussion

3.3.1. Growth response

Nitrogen and water level had no significant interaction effect on the growth parameters measured. Table 4 shows the average LAI, plant height and number of leaves per plant (means across treatments) taken from seven months after planting until 11 months after planting, which was four months after water treatment application. The average LAI, plant height and number of leaves per plant decreased over the growing period across all water treatments (Table 4). The results in Figure 1, further shows that the LAI for the well watered control (30% ADL) dropped slightly after one month of water treatment application, but thereafter there were no further significant reductions. The well watered control always had a significantly higher LAI throughout the growing period (Figure 3.1). Though the severely stressed treatment (70% ADL) had the lowest LAI throughout, it was not significantly different from the moderately stressed treatment (50% ADL). A number of studies on different crops have also reported a decline in LAI values due to water stress. In their studies, Eiasu et al. (2008, 2009) also found that the LAI of rose-scented geranium was negatively affected by water stress, with a significant decline in LAI between the well watered control and the water stressed treatments.

Laurie et al. (2009) found a large reduction in LAI, of about 64 - 80%, due to reduced irrigation on different sweet potato varieties.

Table 3.4. Averages of plant height, number of leaves and leaf area index of P. sidoides over the growth period. Growth period LAI (m2 leaf area/m2 Plant height (cm) Number of leaves/plant (MAP*) ground area) 7 1.40a - - 8 1.18b 17.40a 173a 9 1.08bc 17.80a 162b 10 0.95cd 16.50b 153c 11 0.88d 14.30c 123d

LSD0.05 0.13 0.65 9.07 *MAP = month after planting. Values with different letters are significantly different from each other. Values represent means across all water treatments. LSD: least significant differences.

2 1.8 a LSD 0.05 = 0.23 1.6 b b

groundarea) 1.4 b

2 - 1.2 bcd bcd def 30 % ADL 1 efg 50 % ADL fg fg 0.8 g g leaf area∙ m 70 % ADL 2 0.6 0.4 0.2

Mean LAI MeanLAI (m 0 8 9 10 11 Growth period (months after planting)

Figure 3.1. Leaf area index of P. sidoides in response to water treatment, over the growing period. LSD: least significant difference, ADL: allowable depletion level.

Plant height was also significantly reduced by water stress, as shown in Figure 3.2. The well watered control had a significantly higher plant height but there were no significant differences observed between the moderately and severely stressed treatments. Mabhaudhi et al. (2011) reported a marginal decrease in plant height of bambara landraces under rainfed conditions 30

when compared to irrigated conditions. This could be due to the lower amount of rain received by crops under rainfed conditions. Alishah et al. (2006) found that in basil an increase in water stress levels resulted in a decrease in plant height. Similar results were also reported for Jatropha curcas (Hedayati et al., 2013).

LSD 0.05 = 1.29 20 ab a ab bc c cd c cd e de 15 f 30% ADL f 50% ADL 10

70% ADL Plant height (cm) 5

0 8 9 10 11 Growth period (months after planting)

Figure 3.2. Plant height of P. sidoides over the growing period in response to water treatment. LSD: least significant difference, ADL: allowable depletion level.

Within the second month of water treatment application (8 months after planting - MAP) there were no significant differences in number of leaves across all the treatments (Figure 3.3). However after three and four months of water treatment application (9 and 10 MAP, respectively) the severely stressed treatment (70% ADL) had a significantly lower number of leaves, while there were no significant differences between the well watered control and the moderately stressed treatments. No significant differences were observed again at four months after water treatment application between all the treatments. The sudden increase in number of leaves, 11 month after planting, was due to the fact that the data was taken a week after irrigation of all water treatments, resulting in plant recovery. Since most of these new leaves were small, the mean LAI did not increase (Figure 3.1). Significant reductions in the number of leaves due to water stress were also reported in other crops such as parsley (Petropoulos et al., 2008), basil (Alishah et al., 2006) and common beans (Ghanbari et al., 2013). According to Munné-Bosch and Alegre (2004) leaf senescence is an adaptive strategy that contributes to plant survival under stress, including water stress. Decreasing of the canopy leaf area through reduced growth, is also another strategy to minimize water loss under water stress conditions 31

(Chaves et al., 2003). Leaf senescence may be sequential, starting gradually from oldest to youngest leaves, and depending on the duration and severity of stress, may allow young leaves to grow once stressful conditions have passed (Munné-Bosch and Alegre, 2004). A decrease in total leaf surface due to water stress induced senescence leading to leaf abscission, was reported on Cichorium intybus (Vandoorne et al., 2012).

200 LSD = 17.74 abc a 0.05 180 ab ab abc

1 bc - 160 bc c 140 d d de 120 e 100 30% ADL 80 50% ADL 60 70% ADL

Number Number leavesof ∙ plant 40 20 0 8 9 10 11 Growth period (months after planting)

Figure 3.3. Number of leaves of P. sidoides over the growing period, in response to water treatment. LSD: least significant difference, ADL: allowable depletion level.

Water treatment also had a significant effect on leaf area, as indicated in Table 3.5. There was a declining trend in leaf area with an increase in water stress. The well watered control had a significantly higher leaf area compared to the severely stressed treatment. The moderately stressed treatment showed no significant difference in leaf area when compared to both the well watered control and severely stressed treatment, respectively. Similar results were reported by Liu et al. (2006) on potatoes, where the full irrigation treatment (irrigating to compensate for the full evapotranspiration water loss) had a significantly higher leaf area compared to the deficit irrigation and partial root drying treatments, with no significant differences observed between the two stress treatments. Karkanis et al., 2011, found the lowest leaf area of velvetleaf plant (Abutilon theophrasti) for the water stressed treatment (50% refill of field capacity refill), with the highest leaf area in the well watered control (100% field capacity refill). In another study on vegetable amaranth, Liu and Stützel (2002) reported a significantly lower leaf area for water stressed plants, where irrigation was withheld for a

certain period, compared to the control which was irrigated to 90% of the water holding capacity of the soil, for all genotypes studied.

Table 3.5. Average leaf area index, leaf area per plant and stomatal conductance of P. sidoides, in response to water treatment.

Treatments Mean LAI Mean leaf area Mean conductance ADL (%) m2 leaf area · m-2 ground area cm2 · plant-1 mmol m-2 s-1 30 1.328a 899.6a 100.5a 50 1.009b 707.9ab 49.04b 70 0.934b 617.3b 36.44c

LSD0.05 0.11 204 9.99 *LSD: least significant difference, ADL: allowable depletion level.

Although nitrogen application had no significant effect on the other parameters, it had an effect on number of leaves at harvest. Figure 3.4 shows that nitrogen at the rate of 100 kg · N · ha-1 had a significantly higher number of leaves than the other treatments. Hussain et al. (2006) reported that the maximum number of branches per plant was found with the application of 90 kg · N · ha-1, on asparagus. Zhu et al. (2009) found that the application of medium amounts of N and P fertilizer on Bupleuri radix, either alone or in combination, increased shoot growth amongst other parameters. Araya et al. (2006) reported that application of organic N at the rate of 100 kg · ha-1 increased fresh herbage and oil yield of rose scented geranium, compared to the control in the first harvest. In their second harvest both inorganic and organic N at the rate of 100 kg · ha-1 increased fresh herbage and oil yield over the control (Araya et al., 2006).

200 1 - 180 LSD 0.05 = 19 a

160 b b b

140

120 Number Number leavesof ∙ plant

100 0 50 100 150 Nitrogen (kg ha-1)

Figure 3.4. Average number of leaves per plant, of P. sidoides, in response to nitrogen level, at harvesting. LSD: least significant difference.

3.3.2. Stomatal conductance

The average stomatal conductance of P. sidoides as affected by water stress at different times over the growth period is shown in Figure 3.5. Stomatal conductance of the well watered control (30% ADL) was always significantly higher than that of the stressed treatments, while the 50% and 70% ADL treatments did not differ significantly from each other in most cases. Nitrogen did not show a significant effect on stomatal conductance. Green and Mitchell (1992) also reported a lack of N-related differences in stomatal response to water stress of loblolly pine seedlings.

160 ) )

1 a

- 140 LSD 0.05 = 16.07 a

s 2 2 120 100 b 30% ADL 80 c c 50% ADL 60 cd de de de 70% ADL 40 ef f

20 f Conductance (mmol Conductance (mmol m 0 8 9 10 11 Growing period (months after planting)

Figure 3.5. Average stomatal conductance of P. sidoides at different times over the growing period, in response to water treatments. LSD: least significant difference, ADL: allowable depletion level.

The stomatal conductance results, when observed across water stress treatments (Table 3.5), showed that it decreased with an increase in the stress. The well watered control had a significantly higher stomatal conductance compared to the other two water treatments. The moderately stressed treatment (50% ADL) also had a significantly higher stomatal conductance than the severely stressed treatment (70% ADL). The results on stomatal conductance in this study are consistent with results of work done on other crops. Eiasu et al. (2012) found that rose-scented geranium plants (Pelargonium capitatum) exposed to water stress had a lower stomatal conductance compared to those irrigated more often. Karkanis et al. (2011) reported that water stress reduced stomatal conductance of velvetleaf by 37 – 89%. All the species studied by Galméz et al. (2007) showed a progressive decline in stomatal conductance as water stress intensified.

The increase in stomatal conductance observed in month 10 and 11, could have been due to a decrease in vapor pressure deficits (VPD) in March and April 2013 (Table 3.3). Increases in VPD between leaf and air results in partial closure of the stomata, thus decreasing stomatal conductance so as to prevent excessive dehydration and physiological damage (Oren et al., 1999, Ocheltree et al., 2014). Sweet pepper plants grown under low VPD consistently maintained a higher stomatal conductance compared to plants grown at ambient and high level

VPD (Zabri and Burrage, 1998). Similar results were reported by Comstock and Ehleringer (1993) in their study on common beans; and Dai et al. (1992) on castor bean.

The higher stomatal conductance observed for the well watered control was the result of fully open stomata on both the abaxial and adaxial side of the leaves (Figure 3.6.A and B). The stomata on the moderately stressed samples were opened on the abaxial side and partially closed on the adaxial side of the leaf sample. The lowest stomatal conductance of the severely stressed plants was due to the stomata that were partially closed on the abaxial side, to fully closed on the adaxial side, as was observed under the electron microscope.

Although stomatal regulation in response to water stress has been a controversial issue for long, it has been recognized that stomatal closure results in a limiting conductance, controlling the flow of water through the plant (Comstock and Mencuccini, 1998). The mentioned study suggested a simple threshold model where stomatal closure is triggered as leaf water potential reaches a critical stress level. Stomatal closure is thus amongst the earliest responses to water stress, protecting the plants from extensive water loss (Chaves et al., 2003).

30% ADL 50% ADL 70% ADL

Figure 3.6.A. P. sidoides leaf samples showing stomatal pores as observed under an electron microscope. O = old leaves, M = mature leaves and Y = young leaves. B = abaxial side and T = adaxial side

Figure 3.6.B. P. sidoides leaf samples showing stomatal pores as observed under an electron microscope. O = old leaves, M = mature leaves and Y = young leaves. B = abaxial side and T = adaxial side.

3.4. Conclusions

There was no significant interaction effect between nitrogen and water level for all the parameters measured in this study. Water stress significantly decreased stomatal conductance. Closing of the stomata is a physiological mechanism employed by plants to cope with water stress. However because stomata are the pathway for both water and CO2 exchange with the atmosphere, this mechanism has a negative effect on photosynthesis, and therefore on plant growth. Microscopic observations confirmed that P. sidoides, like most other plant species, respond to water stress by closing their stomata. It was also observed that water stress resulted in closing of the stomata on the adaxial side of the leaves first, followed by closing of those on the abaxial side.

Morphologically, plants respond to water stress by leaf senescence, smaller canopy and smaller leaves, amongst others. P. sidoides showed similar response with reductions in LAI, plant height and leaf area per plant. These observed morphological responses probably resulted from reduced photosynthetic rate, since CO2 uptake was decreased by closing of the stomata under stress conditions. Nitrogen level had a significant effect on number of leaves per plant, but not on leaf area, which means that although more leaves were stimulated, they were not bigger in size and therefore did not result in higher LAI. The study presented the first report on response of P. sidoides to water stress and nitrogen levels, which could be important in the establishment of nitrogen and water management guidelines for cultivation of P. sidoides.

References

Alishah, H. M., Heidari, R., Hassani, A. and Dizaji, A.A. 2006. Effect of water stress on some morphological and biochemical characteristics of purple basil (Ocimum basilicum). J. Biol. Sci 6: 763 – 767

Alva, A.K., Paramasivam, S., Fares, A., Delgado, J.A., Mattos Jr, D. and Sajwan, K. 2006. Nitrogen and irrigation management practices to improve nitrogen uptake efficiency and minimize leaching losses. J. Crop Improv. 15: 369 – 420

Araya, H.T., Soundy, P., Steyn, J.M., Teubes, C., Learmonth, R.A. and Mojela, N. 2006. Response of herbage yield, essential oil yield and composition of South African rose- scented geranium (Pelargonium sp.) to conventional and organic nitrogen. J. Essent. Oil Res. 18: 111 – 115

Ayana, M. 2011. Deficit irrigation practices as alternative means of improving water use efficiencies in irrigated agriculture: Case study of maize crop at Arba Minch, Ethiopia. Afr. J. Agric. Res. 6: 226 – 235

Bilibio, C., Carvalho, J.A., Hensel, O. and Richter, U. 2011. Effect of different levels of water deficit on rapeseed (Brassica napus L.) crop. Ciênc. Agrotec. 35: 672 – 684

Bojović, B. and Marković, A. 2009. Correlation between nitrogen and chlorophyll content in wheat (Triticum aestivum L.). Kragujevac J. Sci. 31: 69 – 74

Bota, J., Medrano and H., Flexas, J. 2004. Is photosynthesis limited to decreased Rubisco Activity and RuBP content under progressive water stress? New phytol. 162: 671 – 681

Brendler, T. and van Wyk, B.-E. 2008. A historical, scientific and commercial perspective on

the use of Pelargonium sidoides (Geraniaceae). J. Ethnopharmacol. 119: 420 – 433

Bunce, J.A. 1996. Does transpiration control stomatal responses to water vapour pressure deficit? Plant Cell Environ. 19: 131 – 135

Chaves, M.M., Pereira, J.S., Maroco, J., Rodrigues, M.L., Ricardo, C.P.P., Osório, M.L., Carvalho, I., Faria, T. and Pinheiro, C. 2002. How plants cope with water stress in the field. Photosynthesis and growth. Ann. Bot. 89: 907 – 916

Chaves, M.M., Maroco, J.P. and Pereira, J.S. 2003. Understanding plant response to drought – from genes to the whole plant. Funct. Plant Biol. 30: 239 – 264

Comstock, J and Ehleringer, J. 1993. Stomatal response to humidity in xommon bean (Phaseolus vulgaris): Implications for maximum transpiration rate, water use efficiency and productivity. Aust. J. Plant Physiol. 20: 669 – 691

Comstock, J. and Mencuccini, M. 1998. Control of stomatal conductance by leaf water potential in Hymenoclea salsola (T. & G.), a desert subshrub. Plant Cell Environ. 21: 1029 – 1038

Dai, Z., Edwards, G.E. and Ku, M.S.B. 1992. Control of photosynthesis and stomatal conductance in Ricinus communis L. (castor bean) by leaf to air vapor pressure deficit. Plant Physiol. 99: 1426 – 1434

Eiasu, B.K., Steyn, J.M. and Soundy, P. 2008. Growth and essential oil yield of rose-scented geranium (Pelargonium capitatum x P. radens ‘Rose’) as affected by withholding irrigation at different times during regrowth. New Zeal. J. Crop Hort. Sci. 36: 285 – 294

Eiasu, B.K., Steyn, J.M. and Soundy, P. 2009. Rose-scented geranium (Pelargonium capitatum X P. radens) growth and essential oil yield response to different soil water depletion regimes. Agric. Water Manage. 96: 991 – 1000

Eiasu, B.K., Steyn, J.M. and Soundy, P. 2012. Physiomorphological response of rose-scented geranium (Pelargonium spp.) to irrigation frequency. S. Afr. J. Bot. 78: 96 – 103

Flexas, J., Ribas-Carbó, M., Bota, J., Galmé, J., Henkle, M., Martínez-Cañellas, S. and Medrano, H. 2006. Decreased Rubisco activity during water stress is not induced by decreased relative water content but related to conditions of low stomatal conductance

and chloroplast CO2 concentration. New phytol. 172: 73 – 82

Galméz, J., Medrano, H. and Flexas, J. 2007. Photosynthetic limitation in response to water stress and recovery in Mediterranean plants with different growth forms. New Phytol. 175: 81 – 93

Ghanbari, A.A., Shakiba, M.R., Toorchi, M. and Rajab, C. 2013. Morpho-physiological responses of common bean leaf to water deficit stress. Euro. J. Exp. Bio. 3: 487 – 492

Green, T.H. and Mitchell, R.J. 1992. Effects of nitrogen on the response of loblolly pine to water stress. I. Photosynthesis and stomatal conductance. New Phytol. 122: 627 – 633

Hedayati, A., Alizadeh, O., Sharafzadeh, S., Abbasi, M. and Azarpanah, A. 2013. Physiological and morphological responses of biodiesel plant (Jatropha curcas L.) to water stress conditions. Intl. J. Crop Sci. 5: 695 – 703

Helfer M, Koppensteiner H, Schneider M, Rebensburg S, Forcisi S, Müller, C., Schmitt- Kopplin, P., Schindler, M., Brack-Werner, R. 2014. The Root Extract of the Medicinal Plant Pelargonium sidoides is a Potent HIV-1 Attachment Inhibitor. PLOS ONE 9, e87487. doi:10.1371/journal.pone.0087487

Hokmalipour, S. and Darbandi, M.H. 2011. Effects of nitrogen fertilizer on chlorophyll content and other leaf indicate in three cultivars of maize (Zea mays L.). World Appl. Sci. J. 15: 1806 – 1811

Hussain, A., Anjum, F., Rab, A. and Sajid, M. 2006. Effect of nitrogen on the growth and yield of asparagus (Asparagus officinalis). J. Agric. Biol. Sci. 1: 41 – 47

Kara, B. and Mujdeci, M. 2010. Influence of late-season nitrogen application on chlorophyll content and leaf area index in wheat. Sci. Res. Essays 5: 2299 – 2303

Karkanis, A., Bilalis, D. and Efthimiadou, A. 2011. Architectural plasticity, photosynthesis and growth responses of velvetleaf (Abutilon theophrasti Medicus) plants to water stress in a semi-arid environment. Aust. J. Crop Sci. 5 (4): 369 – 374

Kumar, A., Signh, D.P. and Singh, P. 1994. Influence of water stress on photosynthesis, transpiration, water-use efficiency and yield of Brassica juncea L. Field Crop Res. 37: 95 – 101

Laurie, R.N., Du Plooy, C.P. and Laurie, S.M. 2009. Effect of moisture stress on growth and performance of orange fleshed sweetpotato varieties. Afric. Crop Sci. Conf. Proc. 9: 235 – 239

Lee, Y.J., Yang, C.M., Chang, K.W. and Shen, Y. 2011. Effects of nitrogen status on leaf anatomy, chlorophyll content and canopy reflectance of paddy rice. Bot. Stud. 52: 295 – 303

Lewu, F.B., Grierson, D.S. and Afolayan. A.J. 2006. Clonal propagation of Pelargonium

sidoides: A threatened medicinal plant of South Africa. Afr. J. Biotechnol. 2 (5): 123

– 125

Lewu, F.B., Adebola, P.O. and Afolayan. A.J. 2007. Commercial harvesting of Pelargonium

sidoides in the Eastern Cape, South Africa: Striking a balance between resource

conservation and rural livelihoods. J. Arid Environ. 70: 380 – 388

Liu, F. and Stützel, H. 2002. Leaf expansion, stomatal conductance and transpiration of vegetable amaranth (Amaranthus sp.) in response to soil drying. J. Amer. Soc. Hort. Sci. 127: 878 – 883

Liu, F., Shahnazari, A., Andersen, M.A., Jacobson, S.E. and Jensen, C.R. 2006. Effects of deficit irrigation (DI) and partial root drying (PRD) on gas exchange, biomass partitioning and water use efficiency in potato. Sci. Hort. 109: 113 – 117

Mabhaudhi, T., Modi, A.T. and Beletse, Y.G. 2011. Growth response of a Bambara groundnut landrace to water stress. Afric. Crop Sci. Conf. Proc. 10: 97 – 102

Motsa, N.M. 2006. Essential oil yield and composition of rose-scented geranium (Pelargonium sp) as influenced by harvesting frequency and plant shoot age. MSc thesis. University of Pretoria, South Africa.

Munné-Bosch, S. and Alegre, L. 2004. Die and let live: leaf senescence contributes to plant survival under drought stress. Funct. Plant Biol. 31: 203 - 216

Ocheltree, T.W., Nippert, J.B. and Prasad, P.V.V. 2014. Stomatal response to changes in vapour pressure deficit reflect tissue-specific differences in hydraulic conductance. Plant Cell Environ. 37: 132 – 139

Oren, R., Sperry, J.S., Katul, G.G., Pataki, D.E., Ewers, B.E., Phillips, N. and Schäfer, K.V.R. 1999. Survey and synthesis of intra and interspecific variation in stomatal conductivity to vapour pressure deficit. Plant Cell Environ. 22: 1515 – 1526

Payne, R.W., Murray, D.A., Harding, S.A., Baird, D.B. and Soutar, D.M. 2008. GenStat for Windows®. 11th ed. Hemel Hempstead (UK): VSN International

Petropoulos, S.A., Daferera, D., Polissiou, M.G. and Passam, H.C. 2008. The effect of water deficit stress on the growth, yield and composition of essential oils of parsley. Sci. Hort. 115: 393 – 397

Schapendonk, A.H.C.M., Spitters, C.J.T. and Groot, P.J. 1989. Effects of water stress on photosynthesis and chlorophyll fluorescence of five potato cultivars. Potato Res. 32: 17 – 32

Shangguan, Z.P., Shao, M.A. and Dyckmans, J. 2000. Nitrogen nutrition and water stress effects on leaf photosynthetic gas exchange and water use efficiency in winter wheat. Environ. Exper. Bot. 44: 141 – 149

Shenkut, A., Tesfaye, K. and Abegaz, F. 2013. Determination of water requirement and crop coefficient for sorghum (Sorghum bicolor L.) at Melkassa, Ethopia. Sci. Technol. Arts Res. J. 2: 16 – 24

Snedecor, G.W. and Cochran, W.G. 1980. Statistical methods. 7th ed. Iowa State University Press

Streck, N.A. 2003. Stomatal response to water vapour pressure deficit: an unresolved issue. Rev. Bras. de Agrociẽncia 9: 317 – 322

Tricker, P.J., Trewin, H., Kull, O., Clarkson, G.J.J., Eensalu, E., Tallis, M.J., Colella, A., Doncaster, C.P., Sabatti, M. and Taylor, G. 2005. Stomatal conductance and not stomatal density determines the long-term reduction in leaf transpiration of poplar in elevated

CO2. Oecologia 143: 652 – 660

Vandoorne, B., Mathieu, A.-S., Van den Ende, W., Vergauwen, R., Périlleux, C., Javaux, M. and Lutts, S. 2012. Water stress drastically reduces root growth and inulin yield in Cichrium intybus (var. sativum) independently of photosynthesis. J. Exper. Bot. 63: 4359 - 4373

Zabri, A.W. and Burrage, S.W. 1998. The effects of vapour pressure defict (VPD) and

enrichment with CO2 on photosynthesis, stomatal conductance, transpiration rate and water use efficiency (WUE) of sweet pepper (Capsicum anuum L.) grown by NFT. Acta Hort. 458: 351 – 356

Zhu, Z., Liang, Z., Han, R. and Dong, J. 2009. Growth and saikosaponin production of the medicinal herb Buplerum Chinese DC. under different levels of nitrogen and phosphorus. Ind. Crops Prod. 29: 96 – 101

CHAPTER 4 YIELD RESPONSE OF PELARGONIUM SIDOIDES DC. TO WATER AND NITROGEN MANAGEMENT

Abstract

Pelargonium sidoides is an important traditional medicine in South Africa whose successful commercialization has resulted in intensified harvesting of wild populations. Information on cultivation practices such as nitrogen and irrigation requirements of P. sidoides is limited or non-existent. The aim of the study was to investigate the effect of water and nitrogen level on yield components of P. sidoides. The study was conducted under a rainshelter as a randomized complete block design with three irrigation levels, four nitrogen levels and three replicates. The irrigation treatments were 30% allowable depletion level (ADL) (well watered treatment), 50% ADL (moderately stressed treatment) and 70% ADL (severely stressed treatment), while the nitrogen levels were 0, 50, 100 and 150 kg · N · ha-1. Nitrogen and water levels had no significant interaction effect on measured parameters. The well watered treatment had a significantly higher total biomass, fresh and dry root yield and nitrogen use efficiency compared to the water stressed treatments. Water use efficiency was significantly lower under well watered conditions. Water stress significantly increased nitrogen content and chlorophyll content of P. sidoides plants and leaves, respectively. Nitrogen levels of 50 and 100 kg · N · ha-1 resulted in a significantly higher total biomass compared to the control. Plant nitrogen content was increased by the application of nitrogen but the nitrogen use efficiency was significantly reduced by nitrogen application. The study further showed that irrigation can increase root yield of P. sidoides significantly while N levels did not have a significant effect on root yield. The increase in total biomass with increased N levels in this study, could be an unnecessary expense as it did not translate into an economic benefit. This could have been due to almost sufficient residual N in the soil and the results could be different in a very poor soil. This report provides important information on irrigation and nitrogen management for growers; and could have a significant impact on how farmers manage conditions for cultivation of P. sidoides.

Keywords: water stress, nitrogen, allowable depletion level, medicinal plant

4.1. Introduction

Agriculture is the major user of water resources in many regions of the world. With increasing aridity and a growing world population, water will become an even scarcer commodity with significant uncertainty of supply and quality, in the near future (Pereira et al., 2002; Chaves et al., 2003; Fereres and Soriano, 2007). Understanding the behaviour of every crop to the amount of water applied is necessary to determine the effect that lack or excess of water may have on production, thus enabling appropriate irrigation management (Bilibio et al., 2011).

Irrigation, a major contributor to agricultural productivity, can increase land productivity by about 127% (Eiasu et al., 2009). In a drought prone country like South Africa, a shift in the allocation of water resources to non-agricultural economic sectors due to competition of use, may impose restrictions on agricultural business unless innovative irrigation management for each crop is introduced (Eiasu et al., 2009). Irrigation water management will have to be carried out more efficiently in future, with the aim of saving water while maximising productivity (Fereres and Soriano, 2007).

Deficit irrigation is becoming an important strategy to reduce agricultural water use in arid and semi-arid regions. It involves deliberate under irrigation of crops or irrigating below evapotranspiration requirements of a crop, to reduce water consumption while minimizing adverse effects of extreme water stress on yield (Pereira et al., 2002; Fereres and Soriano, 2007; Ayana, 2011). This practice requires good knowledge of the response of different crops to water deficits. Deficit irrigation does not always decrease yield and properly applied deficit irrigation may even increase crop yield (Bilibio et al, 2011). Fereres and Soriano (2007) mentioned that information on the effect of deficit irrigation on root crops is limited, given the difficulties in quantifying root biomass under field conditions.

Water use efficiency (WUE), which is explained biologically as carbohydrates formed through photosynthesis from CO2, sunlight and water per unit of transpiration, can be simply described as crop yield per unit of water use (Howell, 2001). It can be generally defined as the ratio of crop yield to water used to produce the yield (Zhang et al., 2004), simply put as yield divided by water-use (Blum, 2009). Water-use is the total amount of water applied to grow the crop during the season and it includes transpiration, evaporation, run-off and deep drainage (Hunt and Kirkegaard, 2009). 47

Any stress involving nitrogen deficiencies and water shortages adversely affects the chlorophyll content of plants (Schlemmer et al., 2005). The utilization of nitrogen is coupled to that of water reaching the root zone, since plants take up inorganic nitrogen contained in water absorbed from the soil solution through their root system (Alva et al., 2006). Under deficient water supply conditions, root growth, physiological activity of roots and shoot growth can be constrained, indirectly leading to N-deficiency (Aujla et al., 2007). Restriction of plant development by low levels of external factors such as water, amongst others, causes the internal concentration of nitrogen to rise (Barker and Bryson, 2007). Leaching and plant uptake of nitrogen are therefore dependent on the available soil water, as the transport medium (Matzner and Richards, 1996; Everard et al., 2010). High irrigation levels have been reported to reduce

NO3 levels by leaching nitrogen out from the soil, while low irrigation levels provide insufficient water for N mineralization and for dissolving NO3 (Abdel-Mawly, 2004).

Since nitrogen input is directly linked with irrigation, the most appropriate and cost effective management strategy would therefore be to integrate irrigation and nutrient inputs, especially nitrogen, as the two cannot be managed independently. Water stress tolerant plants growing in climates with seasonal water stress appear to maintain their root and nutrient uptake capacity during low water availability periods; and this may be very important for the continuation of plant growth as water availability declines (Matzner and Richards, 1996).

Over application of nitrogen fertilizers has caused many environmental pollution problems (Lee et al., 2011) and can lead to decreased yields due to luxury consumption (Alva et al., 2006). Nitrogen use efficiency (NUE) is defined as a proportion of all the nitrogen inputs that are removed in the harvested crop biomass, contained in recycled crop residues and incorporated into the soil organic matter and inorganic N pools (Cassman et al., 2002). Nitrogen source can have large effects on total nitrogen concentration in plants whereby plants grown on ammonium nutrition can have twice the nitrogen concentrations as plants grown on nitrate nutrition (Barker and Bryson, 2007).

Pelargonium sidoides is one of several geophytic species that are important traditional medicines in South Africa (Lewu et al., 2006). The tuberous roots of the species are the raw material for a phytomedicine used to treat acute bronchitis and infections of the upper respiratory tract. Over exploitation of the species, for local traditional use and international

pharmaceutical companies has led to declines in wild populations. It has been reported that P. sidoides growing naturally under low rainfall conditions exhibited increased medicinal value, compared to the one growing in high rainfall conditions (White et al., 2008). However, little is known about cultivation of this species especially the effect of soil water and N levels on its growth and yield. The study was thus undertaken to determine the effect of water and nitrogen level on yield of P. sidoides.

4.2. Materials and methods For details on site description, plant material, trial design and treatment application please refer to Chapter 3.

4.2.1. Data collection and statistical analysis Leaf chlorophyll content was measured on a fully matured leaf, with a chlorophyll content meter (SPAD 502 plus, Konica Minolta, Japan) at harvesting. Three plants per plot which represented three replicates were used for this measurement. Total biomass and root yield were determined by weighing the whole plant first and then roots only, on a field scale (Platform digital scale, W113, Richter scale) after harvesting, where six plants per plot were measured. The roots were then oven dried (Economy oven, 620 digital, Labotec) at 70 °C until constant mass, to obtain dry root yield.

A plant sample from each treatment plot was then subjected to nitrogen content analysis by adapting established methods of Jimenez and Ladha (1993); and Matejovic (1995). The samples were finely milled into powder and approximately 8 to 14 mg per sample were weighed into a tin foil container for N determinations on a Carlo Erba NA 1500 Nitrogen/Carbon/Sulphur Analyzer. A dry oxidation method, known as the Dumas method was used, where the sample and tin container were ignited at high temperature (1020°C) in oxygen, on a chrome oxide catalyst, to produce carbon dioxide, nitrogen gas, oxides of nitrogen and other oxides. The gases were then passed through silvered cobalt oxide, then a column of copper at 540°C, which reduces the oxides of nitrogen to nitrogen gas, removing the excess free oxygen. After the removal of water vapour and carbon dioxide, the N2 gas was finally separated from any traces of other gases by gas chromatography using a helium carrier gas and detected by a thermal conductivity detector. The instrument was calibrated using a pure organic compound of known composition and the compound chosen for our calibration standard was

the ethyl ester of 4-Aminobenzoic acid, which contains 8.48% N. PeakNet software (Dionex Corporation, 1998) with an external A/D interface (UI20 Universal interface, Dionex) was used for data collection, peak integration, calibration and computation of concentrations. Plant nitrogen content was determined on dry plant material, thus the values are on a dry weight basis. Total N taken up per plot was determined using the formula:

Total N uptake = (Ng/X) x Y

Where Ng is nitrogen content (grams) of the plant sample, X is the mass of the plant (grams) and Y is the total mass of all the other plants from the same plot.

NUE was estimated as amount of dry matter produced per kg nitrogen applied (Fageria and Baligar, 2005). WUE was calculated according to Hunt and Kirkegaard (2009) and expressed as g · mm -1. In this study the roots are the harvested parts and thus root yield, on a dry mass basis, was used in both the calculations and therefore both are expressed as NUEroot and

WUEroot.

Data were subjected to analysis of variance (ANOVA) using GenStat® version 11.1. Treatment means were separated using Fisher’s protected T-test for least significant differences (LSD) at 5% level of significance.

4.3. Results and discussion

4.3.1. Soil water deficit and water use Figure 4.1 shows the soil water deficits allowed, per water treatment, before irrigation was initiated. The amount of water applied over the treatment period was 466 mm for the well watered treatment (30% ADL), 307 mm for the moderately stressed treatment (50% ADL) and 256 mm for the severely stressed treatment (70% ADL). The total relative evapotranspirative demand (ET0) for the treatment period was 663.3 mm (see Table 3.2). The well watered treatment in this study recorded the highest water usage which is in agreement with research done by Eiasu et al. (2009) on rose-scented geranium. Darwish et al. (2006) also recorded the highest water usage, of 510.6 mm, for the well watered control (irrigating to 100% ET), followed by 407.3 mm for the moderate deficit irrigation (irrigating to 80% ET) and 313.6 mm for the severe deficit irrigation (irrigating to 60% ET). 50

30% ADL

58.0 50% ADL

70% ADL

48.0 (mm) 38.0

28.0

18.0 Soil water deficit Soilwater 8.0

-2.0

Date of measurement

Figure 4.1. Changes in soil water deficits (top 400 mm root zone) for the well watered (30% ADL), moderately stressed (50% ADL) and severely stressed (70% ADL) treatments.

4.3.2. Total fresh biomass yield There was no significant interaction between nitrogen and water stress level for all parameters measured. N rate had a significant effect on total biomass where there was an increase in total biomass from 0 to 50 kg · N · ha-1 (Figure 4.2.). Thereafter a slight decrease was observed at 100 and again at 150 kg · N · ha-1, although it was not significantly different from total biomass produced with the 50 kg · N · ha-1. Hussain et al. (2006) reported that a nitrogen rate of 90 kg · N · ha-1 increased plant weight of asparagus and attributed this increase to an increase in number of branches and plant height, which enhanced photosynthetic activity. The highest total biomass for palmarosa, an essential oil crop, was produced by applying 80 kg · N · ha-1 under rainfed conditions (Rao, 2001). A rate of 160 kg · N · ha-1 significantly increased herbage yield of P. graveolens, beyond which there were no significant yield improvements (Ram et al., 2003). Araya et al. (2006) found an increase in fresh herbage yield over the control at 100 kg · N · ha-1 beyond which there were no significant increases in yield for rose-scented geranium (Pelargonium capitatum).

35 LSD = 4.23 30 0.05 a a

25 a

) 1

- 20 b b ha

· 15 (t 10

5 Totalfresh biomass yield 0 0N 50N 100N 150N Nitrogen (kg · ha -1)

Figure 4.2. Total fresh biomass yield of P. sidoides in response to nitrogen level.

Water treatments also had a significant effect on total fresh biomass yield. Figure 4.3 shows that the well watered treatment had a significantly higher total biomass yield than the two water stressed treatments. There was no significant difference between the moderately and severely stressed treatments. Similar results were reported previously on other crops. Eiasu et al. (2009) reported that the well watered treatment, referred to as lower maximum allowable depletion (MAD) level in that study, resulted in better herbage yield of rose-scented geranium. Karkanis et al. (2011) also found that velvetleaf plants produced the lowest aboveground biomass under water stress. Moderate water stress, at 50 % MAD, and severe water stress, at 80% MAD, resulted in significant decreases in above ground fresh yield of Thymus daenensis, when compared to the non- stressed control (20% MAD) (Bahreininejad et al., 2013). The same study found no significant differences between the moderate and severe water stress treatment on above ground fresh yield, in the second year.

35 30 a LSD 0.05 = 3.67

25 )

1 b

- 20 b ha

· · 15 (t 10 5

Totalfresh biomass yield 0 30% 50% 70% Water depletion (% ADL)

Figure 4.3. The effect of soil water level on total fresh biomass yield of P. sidoides.

In the first year of harvesting P. sidoides from the wild, Lewu et al. (2007) found a fresh biomass yield of 5.7 kg · plot-1 of 20 m2 with a mean population of 23.6 plants, which translates to 11 800 plants · ha-1. This yields to about 0.24 kg · plant-1 or 2.85 t · ha-1. In our study the severe water stress treatment (70% ADL) resulted in a similar yield of 0.24 kg · plant-1. However, the increase in plant population from 11 800 plants · ha-1 as reported in the literature by Lewu et al. (2007) to 66 666 plants · ha-1 in our study, increased yield from 2.85 to 16 t · ha-1 under water stressed conditions. Although the growth period of the wild population was not determined by Lewu et al. (2007), the current study shows that irrigation can increase total biomass yield to more than 26 t · ha-1 when compared to only 2.85 t · ha-1 as reported in the literature under rainfed conditions in the wild.

4.3.3. Root yield Fresh root yield and dry root yield, respectively, followed the same trend as biomass yield where water stress significantly reduced the yield (Table 4.3), while N application had no significant effect on root yield. The well watered treatment resulted in significantly higher values than the water stressed treatments for both fresh and dry root yield, with no significant differences between the water stressed treatments. Plants exposed to moderate water stress had a significantly lower dry matter content compared to the well watered treatment (Table 4.3). Other researchers reported similar results in different crops. Vandoorne et al. (2012), in their study on root chicory, found that the mean root fresh yield was lowered by water stress, with a decrease of more than 50%, compared to the control. Fresh root yield of plain leafed and turnip rooted parsley exposed to a higher water stress level was significantly decreased in both years of the study (Petropoulos et al., 2008). Darwish et al. (2006) reported that severe deficit irrigation (60% ET) led to a 21% loss in potato fresh yield, due to lowered tuber dry matter production and average mass of the commercial tubers. The dry matter yield of whole plants and individual parts of Chenopodium quinoa plants was significantly higher in the control than under drought conditions (Gonzàlez et al., 2009). In their study, Bahreininejad et al. (2013), found that the dry root yield of Thymus daenensis was significantly higher in the non-stressed control (20% MAD) compared to the moderate and severe water stress treatments (50% and 80% MAD, respectively), while there were no significant differences between the two water stress treatments in the first year. The low prices (R1.60 – R4.50 · kg-1) fetched by harvesters for wet mass of P. sidoides roots as compared to what the middleman is paid (R48 – R120 · kg-1) (Motjotji, 2011) could be due to the initial high moisture content in the roots, in addition

to marketing costs. If the harvesters were to dry the plant material themselves, they would possibly fetch a maximum price of R18.00 · kg-1, given the lower moisture content, but this means the middleman still makes a major profit.

Table 4.1. Fresh, dry root yield and total biomass dry matter content (percentage) of P. sidoides in response to water treatment. Water depletion Fresh root Dry root Average dry matter (% *ADL) yield (t · ha-1) yield (t · ha-1 ) content (%)

30 19.77 a 3.58 a 27.24 a 50 14.52 b 2.51 b 24.58 b 70 14.05 b 2.15 b 25.57 ab

LSD0.05 2.81 0.59 1.93 * LSD = least significant difference, ADL = Allowable depletion level

4.3.4. Chlorophyll and nitrogen content Water stress significantly increased chlorophyll content as shown in Figure 4.4. There was a significant increase from the well watered treatment (30% ADL) to the moderately stressed treatment (50% ADL); however there was no significant difference between the two stress treatments. The nitrogen content (%) in the leaves and in the whole plant (Table 4.4) followed a similar trend, where both the moderately stressed and severely stressed treatments (70% ADL) were not significantly different from each, other but were significantly higher than the well watered treatment. Our results are consistent with findings on other crops. Gonzàlez et al. (2009) reported no significant differences in chlorophyll content of Chenopodium quinoa plants, between the control and drought treatments, but the leaf nitrogen content was significantly higher in the drought treatment compared to the control. Rahimi et al. (2010) also reported that drought stress increased chlorophyll content significantly in Plantago ovata and P. psyllium plants. A decrease in relative water content and leaf water potential, due to water stress, resulted in increased leaf chlorophyll content, by 17 and 37%, for P. ovata and P. psyllium, respectively (Rahimi et al., 2010).

70 LSD 0.05 = 2.45 60 a a 50 b 40 30

SPAD SPAD reading 20 10 0 30 50 70 Water deletion (% ADL)

Figure 4.4. Chlorophyll content (SPAD) of P. sidoides as affected by soil water level.

Table 4.2. Nitrogen content (%) in the leaves and in whole P. sidoides plant in response to water treatments. Water depletion N content (leaves) N content (whole plant) (% *ADL) % 30 2.69 b 1.50 b 50 3.17 a 2.16 a 70 3.21 a 2.26 a

LSD0.05 0.19 0.24 *LSD = least significant difference, ADL = Allowable depletion level

An increase in nitrogen rate (kg · ha-1) resulted in a significant increase in nitrogen content (%) of the whole plant (Figure 4.5). The rate of 150 kg · N · ha-1 resulted in a significantly higher nitrogen content than the control and lowest application. The control, 50 and 100 kg · N · ha-1 did not differ significantly from each other, while the 100 kg · N · ha-1 did also not differ significantly from the 150 kg · N · ha-1 treatment. Similarly, nitrogen application significantly affected nitrogen concentration of potato tubers, where application of N at the rate of 240 -250 kg · ha-1 resulted in significantly higher tuber N concentration (Millard, 1986). Millard and MacKerron (1986) also reported that N application increased nitrogen concentration in leaves and stems in potatoes. The highest nitrogen content for isabgol (Plantago ovata) under medium and low water stress (60% and 80% of FC, respectively) was observed with the highest application of nitrogen at the rate of 180 mg · N · kg-1 soil in pots (Rahimi et al., 2013).

3 2.5 LSD 0.05 = 0.28 a ab 2 b b 1.5 1 0.5

%contentN (wholeplant) 0 0N 50N 100N 150N

Nitrogen (kg · ha -1)

Figure 4.5. Nitrogen content, on a dry mass basis, of whole plants of P. sidoides as affected by nitrogen level.

4.3.5. Nitrogen uptake and Nitrogen use efficiency (NUEroot) The two water stress treatments, moderate (50% ADL) and severe water stress (70% ADL), showed a significantly higher total nitrogen uptake compared to the well watered treatment, as indicated by Figure 4.6. The results are consistent with findings of Scagel et al. (2011) on Rhododenron, where they reported that a decrease in water stress was associated with lower N uptake. Panicum maximum plants undergoing water stress showed an increase in nitrogen concentration (Dias-Filho et al., 1992). Water stress tolerant growing under seasonal water are reported to have the capability of maintaining nutrient uptake during times of low water availability (Matzner and Richards, 1996). Plants are known to increase their root depth under drought conditions so as to extract water from wet deeper soil layer. Several species have shown the ability to extract nutrients from shallow dry soil layers as long as their root system can access deeper wet layers (Matzner and Richards, 1996).

50 )

1 LSD 0.05 = 3.67 - a

40 a · ha ·

30 b 20

10 TotalN uptake (kg 0 30% 50% 70% Water depletion (% ADL)

Figure 4.6. Total nitrogen uptake, on a dry mass basis, of P. sidoides plants exposed to different water levels.

Nitrogen level again had a significant effect on nitrogen use efficiency (NUEroot) where an increase in the application rate resulted in a decrease in NUEroot (Figure 4.7). The 50 kg · N · -1 -1 ha treatment had significantly higher NUEroot than the 100 and 150 kg · N · ha treatment. Badr et al. (2012) also found that an increase in N application rate resulted in a decrease in NUE of potato tubers. Similar results were reported by Colla et al. (2005), Fandika et al. (2010) and Ospina et al. (2014), on potatoes. Increases of nitrogen level from 80, 160 to 200 kg N · ha-1 resulted in significant decreases in NUE of potato tubers, from 26.81 to 13.32 kg yield · kg-1 N (Jamaati-e-Somarin et al., 2010).

Water stress also had an effect on NUEroot where it was observed that the well watered treatment had a significantly higher NUEroot than the two water stress treatments (Figure 4.8). Abdel-Mawly (2004) reported a marked increase in recovery of applied N with an increase in the supply of irrigation water on carrots. Darwish et al. (2006) attributed the lower NUE observed under moderate and severe deficit irrigation in potatoes, to a lower tuber dry matter yield under those conditions. Irrigation treatment had a significant effect NUE of potatoes where the highest value of 176 kg yield · kg-1 N and lowest values of 55 kg yield · kg-1 N, were reported for the well irrigated treatment and severe water deficit treatment, respectively (Badr et al. 2012). Irrigation increased NUE of four potato cultivars studied by Fandika et al. (2010), by 29.6, 26.2, 60 and 104.2%, respectively.

)

1 - 50 a

LSD 0.05 = 3.87 kg N kg

· · 40

- ha

· · 30 b

(kg (kg 20 c root

10 NUE 0 50 100 150

-1 N application (kg · ha )

Figure 4.7. Nitrogen use efficiency of P. sidoides plants, on a dry root mass basis, as influenced by nitrogen level.

) 1 - a

40 LSD 0.05 = 3.87

kg N kg

· · 1 - 30

ha b b · ·

20 (kg (kg

root 10 NUE 0 30% 50% 70% Water depletion (ADL)

Figure 4.8. Nitrogen use efficiency of P. sidoides plants, on a dry root mass basis, as influenced by soil water level.

4.3.6. Water use efficiency (WUEroot) According to Figure 4.9, the well watered treatment showed significantly lower water use efficiency compared to the water stress treatments, which were not significantly different from -1 each other. A decrease in WUEroot, averaging 1.4 g DW · kg H20, in two varieties of sugarbeet under well watered conditions was reported by Rytter (2005). Irrigation water productivity, which is the final yield per unit of applied water, of potato tubers was increased under moderate water stress (Darwish et al., 2006). Lakew et al. (2014) studied, on onion, the total water use efficiency (TWUE), which is the onion yield per mm-1 of total water applied and net water use efficiency (NWUE) is the yield produced per mm-1 of water consumed. The results indicated 58

that under full irrigation level (FIL, 100% ET0) both TWUE and NWUE were significantly decreased, as compared to the mid irrigation level (MIL, 75% ET0) and half irrigation level

(HIL, 50% ET0) (Lakew et al., 2014). The approximate values for TWUE for FIL was 15.45 kg/mm, for the MIL it was 22.41 kg/mm and for the HIL it was 38.24 kg/mm; while for the NWUE they were 16.85 kg/mm for FIL, 28.81 kg/mm for MIL and 40.02 kg/mm for HIL (Lakew et al., 2014). Eiasu et al. (2012) reported an increase in WUE with an increase in irrigation frequency for rose-scented geranium (Pelargonium spp) and this could be attributed to low biomass production under drought conditions (Wu et al., 2008). The difference could be that in their study, Eiasu et al. (2012), the harvested part was leaves while in the current study it was roots.

10 LSD 0.05 = 0.45

) 9 1 a - a

8 b mm

· 7 1 - 6

· ha 5

(kg (kg 4

3 root 2

WUE 1 0 30% 50% 70% Allowable depletion (ADL)

Figure 4.9. Water use efficiency of P. sidoides, on dry root mass basis, in response to different soil water levels.

4.4. Conclusions The well watered treatment (30% ADL) used more water than the moderate (50% ADL) and severe (70% ADL) treatments and that could be attributed to larger canopies and increased water availability. An increase in water stress resulted in an increase in nitrogen uptake, nitrogen content and chlorophyll content. However, the total biomass and root yield was decreased by water stress. This could be attributed to growth inhibition under water stress, to save and redistribute limited resources, which is counter-productive in terms of yield. WUE, based on dry root mass, was increased by water stress meaning that more root mass was produced per mm-1 of irrigation water applied. The total biomass yield increased with 59

increasing nitrogen level until the rate of 100 kg · N · ha-1, while the nitrogen content of whole plant continued to increase with further increases in nitrogen level. The increased N content, of both the leaves and whole plant, under water stress could be due to the accumulation of N in the plants, but it may not have been utilized for plant development due to water stress, resulting in a low NUEroot under the same stress condition. The study showed that irrigation can increase root yield of P. sidoides significantly, while N levels did not have a significant effect on root yield, although an increase in total biomass was observed. The increase in total biomass at 100 kg · N · ha-1 did not relate into an increase in root yield and is not important in cultivation of P. sidoides since the roots are used for medicinal purposes. The increased total biomass, in this study, is therefore an unnecessary expense as it did not translate into an economic benefit for the farmer. The major profit differences between harvesters and the middleman could mean that the growers or harvesters can be advised to semi-process and supply directly to the market.

References

Abdel-Mawly, S.E. 2004. Growth, yield, N uptake and water use efficiency of carrots (Daucus carota L.) plants as influenced by irrigation level and nitrogen fertilization rate. Ass. Univ. Bull. Environ. Res. 7: 111 – 122

Alva, A.K., Paramasivam, S., Fares, A., Delgado, J.A., Mattos, D. and Sajwan, K. 2006. Nitrogen and irrigation management practices to improve nitrogen uptake efficiency and minimize leaching losses. J. Crop Improv. 15: 360 – 420

Aujla, M.S., Thind, H.S and Buttar, G.S. 2007. Fruit yield and water use efficiency of eggplant (Solanum melongema L.) as influenced by different quantities of nitrogen and water applied through drip and furrow irrigation. Sci. Hort. 112: 142 – 148

Ayana, M. 2011. Deficit irrigation practices as alternative means of improving water use efficiencies in irrigated agriculture: Case study of maize crop at Arab Minch, Ethiopia. Afric. J. Agric. Res. 6: 226 – 235

Badr, M.A., El-Tohamy, W.A. and Zaghloul, A.M. 2012. Yield and water use efficiency of potato grown under different irrigation and nitrogen levels in an arid region. Agric. Water Manage. 110: 9 – 15

Bahreininejad, B., Razmjoo, J. and Mirza, M. 2013. Influence of water stress on morpho- physiological and phytochemical traits in Thymus daenensis. Int. J. Plant. Prod. 7: 151 – 166

Barker, A.V. and Bryson, G.M. 2007. Nitrogen, in: Barker, A.V. and Pilbeam, D.J. (Eds), Handbook of plant nutrition. Taylor and Francis Group, New York pp. 22 – 43

Bilibio, C., Carvalho, J.A., Hensel, O. and Richter, U. 2011. Effect of different levesl of water deficit on rapeseed (Brassica napus L.) crop. Ciênc. Agrotec.Lavras. 35: 672 – 684

Blum, A. 2009. Effective use of water (EUW) and not water use efficiency is the target of crop yield improvement under drought stress. Field Crop Res. 112: 119 – 123

Cassman, K.G., Dobermann, A. and Walters, D.T. 2002. Agroecosystems, nitrogen use efficiency and nitrogen management. Ambio. 31: 132 – 140

Chaves, M.M., Maroco, J.P. and Pereira, J.S. 2003. Understanding plant response to drought – from genes to the whole plant. Func. Plant Biol. 30: 239 – 264

Colla, G., Cardarelli, M., Rouphael, Y. and Fiorillo, A. 2005. Response of potatoes to organic fertilizer rate: yield, tuber quality and nutrient uptake. Acta Hort. 684: 73 – 77

Darwish, T.M., Atallah, T.W., Hajhasan, S. and Haidar, A. 2006. Nitrogen and water use efficiency of fertigated processing potato. Agric. Water Manage. 85: 95 – 104

Dias-Filho, M.B., Corsi, M. and Cusato, S. 1992. Concentration, uptake and use efficiency of N, P and K in Panicum maximum Jacq. Cv. Tobiaotã under water stress. Pesq. Agropec. Bras. 27: 381 – 387

Dionex Corporation, 1998. Peaknet software user’s guide. Revision 08. Doc. No. 034914

Eiasu, B.K, Steyn, J.M and Soundy, P. 2009. Rose-scented geranium (Pelargonium capitatum x P.radens) growth and essential oil yield response to different soil water depletion regimes. Agric. Water Manage. 96: 991 – 1000

Eiasu, B.K, Steyn, J.M and Soundy, P. 2012. Physiomorphological response of rose-scented geranium (Pelargonium spp.) to irrigation frequency. S. Afric. J. Bot. 78: 96 – 103

Everard, K., Seabloom, E.W., Harpole, W.S. and de Mazancourt, C. 2010. Plant water use affects competition for nitrogen: Why drought favors invasive species in California. Americ. Natur. 175: 85 – 97

Evett, S. 2008. Neutron moisture meters. In: Field estimation of soil water content, a practical guide to methods, instrumentation and sensor technology. pp 39 -54. Training course series No. 30. International atomic energy agency, Vienna, Austria.

Fageria, N.K. and Baligar, V.C.2005. Enhancing nitrogen use efficiency in crop plants. Adv. Agron. 88: 97 – 185

Fandika, I.R., Kemp, P.D., Millner, J.P. and Horne, D.J. 2010. Water and nitrogen use efficiency in modern and Maori potato cultivars. Agron. New Zeal. 40: 159 – 169

Fereres, E. and Soriano, M.A. 2007. Deficit irrigation for reducing agricultural water use. J. Exper. Bot. 58: 147 – 159

Gonzàlez, J.A., Gallardo, M., Hilal, M., Rosa, M. and Prado, F.E. 2009. Physiological responses of quinoa (Chenopodium quinoa Willd.) to drought and waterlogging stresses: dry matter partitioning. Bot. Stud. 50: 35 – 42

Howell, T.A. 2001. Enhancing water use efficiency in irrigated agriculture. Agron. J. 93: 281 – 289

Hunt, J. and Kirkegaard, J. 2009. Benchmarking water-use efficiency – a nationally consistent but regionally relevant approach. Grains Research and Development Centre, Water Use Efficiency Newsletter 1: 2 – 5. www.grdc.com.au. Accessed on 31st October 2010

Hussain, A., Anjum, F., Rab, A.and Sajid, M. 2006. Effect of nitrogen on the growth and yield of asparagus (Asparagus officinalis). J. Agric. Biol. Sci. 1: 41 – 47

Jamaati-e-Somarin, S., Zabihi-e-Mahmoodabad, R. and Yari, A. 2010. Response of agronomical, physiological, apparent recovery nitrogen use efficiency and yield of potato tuber (Solanum tuberosum L.), to nitrogen and plant density. American- Eurasian J. Agric & Environ. Sci. 9: 16 – 21

Jimenez, R.R. and Ladha, J.K. 1993. Automated elemental analysis: A rapid and reliable but expensive measurement of total carbon and nitrogen in plant and soil samples. Commun. Soil Sci. Plant Anal. 24: 1897 – 1924

Karkanis, A., Bilalis, D. and Efthimiadou, A. 2011. Architectural plasticity, photosynthesis and growth response of velvetleaf (Abutilon theophrasti Medicus) plants to water stress in semi- arid environment. Australian J. Crop Sci. 5: 369 – 374

Lakew, W.J, Anteneh, B.A and Ayalew, L.T. 2014. Yield and water use efficiency of mulched drip irrigated onion in low land region of Amhara, North Central Ethopia. Uni. J. Agric. Res. 2 : 203 – 210

Lee, Y. J., Yang, C.M., Chang, K.W. and Shen, Y. 2011. Effect of nitrogen status on leaf anatomy, chlorophyll content and canopy reflectance of paddy rice. Bot. Studies 52: 295 – 303

Lewu, F.B., Grierson, D.S. and Afolayan, A.J. 2006. The leaves of Pelargonium sidoides may substitute for its roots in the treatment of bacterial infections. Biol. Conserv. 128: 582 – 584

Lewu, F.B., Adebola, P.O. and Afolayan, A.J. 2007. Commercial harvesting of Pelargonium sidoides in the Eastern Cape, South Africa: Striking a balance between resource conservation and rural development. J. Arid Environ. 70: 380 – 388 77: 193 – 201

Matejovic, I. 1995. Total nitrogen in plant material determined by means of dry combustion: A possible alternative to determination by Kjedahl digestion. Commun. Soil Sci. Plant Anal. 26: 2217 – 2229

Matzner, S.L. and Richards, H.R. 1996. Sagebrush (Artemisia tridentata Nutt.) roots maintain nutrient uptake capacity under water stress. J. Exper. Bot. 47: 1045 - 1056

Millard, P. 1986. The nitrogen content of potato (Solanum tuberosum L.) tubers in realtion to nitrogen application – the effect on amino acid composition and yields. J. Sci. Food Agric. 37: 107 – 114

Millard, P. and MacKerron, D.K.L. 1986. The effects of nitrogen application on growth and nitrogen distribution within the potato canopy. Ann. Appl. Biol. 109: 427 – 437

Ospina, C.A., Van Bueren, E.T.L., Allefs, J.J.H.M., Engel, B., Vand der Putten, P.E.L., Van der Linde, C.G. and Struik, P.C. 2014. Diversity of crop development traits and nitrogen use efficiency among potato cultivars grown under contrasting nitrogen regimes. Euphytica, 199: 13 – 29

Pereira, L.S., Oweis, T. and Zairi, Abdelaziz. 2002. Irrigation management under water scarcity. Agric. Water Manage. 57: 175 – 206

Petropoulos, S.A., Daferera, D., Polissiou, M.G. and Passam, H.C. 2008. The effect of water deficit on the growth, yield and composition of essential oils of parsley. Sci Hort. 115: 393 – 397

Rahimi, A., Hosseini, S.M., Pooryoosef, M. and Fateh, I. 2010. Variation of leaf water potential, relative water content and SPAD under gradual drought stress and stress recovery in two medicinal species of Plantago ovata and P. psyllium. Plant Ecophysiol. 2: 53 – 60

Rahimi, A., Sayadi, F., Dashti, H. and Pour, A.T.2013. Effects of water use and nitrogen supply on growth, water-use efficiency and mucilage yield of isabgol (Plantago ovata Forsk). J. Soil Sci. Plant Nutri. 12: 341 – 354

Ram, M., Ram, D. and Roy, S.K. 2003. Influence of an organic mulching on fertilizer nitrogen use efficiency and herb and essential oil yields in geranium (Pelargonium graveolens). Biores. Technol. 87: 273 – 278

Rao, B.R.R. 2001.Biomass and essential oil yields of rainfed palmarosa (Cymbopogon martinii (Roxb.) Wats. var. motia Burk.) supplied with different levels of organic manure and fertilizer nitrogen in semi-arid tropical climate. Ind. Crops Prod. 14: 171 – 178

Rytter, R.M. 2005. Water use efficiency, carbon isotope discrimination and biomass production of two sugar beet varieties under well watered and dry conditions. J. Agron. Crop Sci. 191: 426 – 438

Scagel, C.F., Bi, G., Fuchigami, L.H. and Regan, R.P. 2011. Effects of irrigation frequency and nitrogen fertilizer rate on water stress, nitrogen uptake and plant growth of container grown rhododendron. HortScience 46: 1598 – 1603

Schlemmer, M.R., Francis, D.D., Shanahan, J.F. and Schepers, J.S. 2005. Remotely measuring chlorophyll content in corn leaves with differing nitrogen levels and relative water content. Agron. J. 97: 106 – 112

White, A.G., Davies-Coleman, M.T. and Ripley, B.S. 2008. Measuring and optimizing umckalin concentration in wild harvested and cultivated Pelargonium sidoides (Geraniaceae). S. Afric. J. Bot. 74: 260 – 267

Wu, F., Bao, W., Li, F. and Wu, N. 2008. Effects of drought stress and N supply on the growth, biomass parttioning and water use efficiency of Sophora davidii seedlings. Eviron. Exp. Bot. 63: 248 – 255

Zhang, Y., Kendy, E., Qiang, Y., Changming, L., Yanjun, S. and Hongyong, S. 2004. Effect of soil water deficit on evapotranspiration, crop yield and water use efficiency in the North China Plain. Agric. Water Manage. 64: 107 – 122

CHAPTER 5

CHEMICAL COMPOSITION OF P. SIDOIDES DC. AS AFFECTED BY SOIL WATER AND NITROGEN LEVEL

Abstract

Many South African medicinal plants, including Pelargonium sidoides, are used in the production of internationally marketed herbal remedies for the treatment of various ailments. The combined traditional and commercial use of P. sidoides has led to the extensive harvest of wild plants in the Eastern Cape and Free State provinces of South Africa and in Lesotho. Currently no relationship has been established between the chemical composition of P. sidoides and production practices such as water and nitrogen supply. The aim of the study was to investigate the effect of soil water and nitrogen levels on chemical composition of P. sidoides. A trial was conducted under a rainshelter, with three predetermined irrigation levels and four different nitrogen levels. The soil profile was irrigated back to field capacity when 30%, 50% and 70% of plant available water was depleted. Nitrogen was applied at rates of 50, 100 and 150 kg · N · ha-1 with a control of no (0) nitrogen. Dried root samples were analyzed for the presence of two standard compounds, scopoletin and esculin, using thin layer chromatography (TLC) and for metabolite profiling using the nuclear magnetic resonance technique (NMR). TLC analysis showed the presence of the two standard compounds in all treatment samples analyzed. The orthogonal partial least square discriminatory analysis (OPLS-DA), which was performed on the NMR spectral data, showed separation between the irrigation treatments, resulting in two clusters representing the well watered treatment and the water stressed treatments. Asparagine, arginine, sucrose, xylose, glucose and citric acid were found to be the compounds associated with the separation. There was no separation of the samples regarding the nitrogen treatments which is indicative of the small effect of nitrogen on the metabolite content of the treatments. The study successfully differentiated the metabolites responsible for variation between well watered and water stressed root samples of P. sidoides. Scopoletin and esculin could not be identified from the NMR profiles as these compounds seemed to be present in lower concentrations than needed for NMR. This could possibly mean that P. sidoides plants need to be cultivated for two seasons or more for the water and nitrogen effects to be observed on the chemical composition, especially marker compounds.

Keywords: water stress, nitrogen, allowable depletion level, secondary metabolites, primary metabolites, NMR, metabolomic profiling.

5.1. Introduction

The performance of plants is greatly influenced by water deficits, in terms of the production and the formation of compounds; with the level and duration of stress being fundamental in the response of essential oil and medicinal plants (Alvarenga et al., 2011, Pirzad et al., 2012). Secondary compounds found in plants are associated with increasing survival, either by coping with unfavourable environmental conditions or by regulating some metabolic processes (Solecka, 1997). The yield and composition of secondary metabolites of certain medicinal plants vary within and between populations as influenced by geographic and environmental factors (White et al., 2008). Plant nutrients, especially nitrogen and water are the most important variables for producing a profitable crop (Schlemmer et al., 2005).

Commercial root extracts of Pelargonium sidoides, an important southern African medicinal plant, have become popular in many countries for the treatment of upper-respiratory tract infections (Seidel and Taylor, 2004, Wittschier et al., 2007, Agbabiak et al., 2008, White et al., 2008, Brendler, 2009). The phytochemical profiling of P. sidoides has indicated a dominating presence of polymeric pro-anthocyanidins and monomeric flavan-3-ols, besides considerable amounts of coumarins (Wittschier et al., 2007). Coumarins have been hypothesised to be produced in response to traumatic injury by plant diseases or during water stress (White et al., 2008, Ojala, 2001). They comprise a very large class of compounds found throughout the plant kingdom and although mainly synthesised in the leaves, they occur at the highest levels in the fruits, followed by the roots, stems and leaves (Ojala, 2001, Lacy and O’Kennedy, 2004, Jain and Joshi, 2012).

The medicinal properties of P. sidoides root extracts are ascribed to eight different coumarins, of which umckalin and 5, 6, 7 – methoxycoumarin are useful marker compounds as they appear to be absent in P. reniforme (Bladt and Wagner, 2007, Brendler and van Wyk, 2008, van Wyk, 2008; White et al., 2008; European medicines agency, 2012). Van Wyk (2008) reported the presence of gallic acids and methyls of gallic acids, flavonoids, flavan-3-ols and phytosterols in the roots. Mativandlela (2005) mentioned that two distinct coumarins; umckalin and its 7- O-methyl ester, together with four other methoxycoumarins and three unique coumarin sulphates were found in P. sidoides but not in P. reniforme. Some studies have suggested that various coumarin glycosides and coumarin sulphates are confined to P. sidoides (Brendler and van Wyk, 2008). The presence of esculin and scopoletin in the root extracts of P. sidoides has

also been reported (Kayser and Kodziej, 1995, Gödecke et al., 2005, Kolodziej, 2007, European Pharmacopoeia 6.0, 2008). Esculetin and scopoletin are some of the most widespread coumarins in nature (Jain and Joshi, 2012). Scopoletin synthesis is post-infectionally activated in plants but can also be triggered by various abiotic stresses (Bourgaud et al., 2006). It also displays radical scavenging properties toward reactive oxygen species and may be involved in the reduction of oxidative stress in plant cells (Bourgaud et al., 2006). Umckalin may not be an active compound of P. sidoides but is used as a marker compound to differentiate between roots of P. sidoides and those of P. reniforme, as reported by Maree and Viljoen (2012). Although P. reniforme does not have umckalin, both the roots of P. sidoides and P. reniforme are traditionally used for the same purposes (Kolodziej, 2002, Mativandlela, 2005, White, 2006, White et al., 2008). P. sidoides has significantly higher yield of coumarins, with umckalin amounting for about 40% of total coumarin content (European medicines agency, 2012) and this could be a possible reason for preference of P. sidoides roots.

Metabolomics is an increasingly important OMICS method which comprises the detection of metabolites present within a biological system under specific conditions, thus providing a phenotypic assessment of the system (Tugizama et al., 2013). Biological important molecules have signals which are characterized by their frequency or chemical shift, intensity, structure and magnetic relaxation properties; all of which reflects the environment of the sample under detection (Krishana et al., 2005).

The use of irrigation in cultivation of medicinal plants can increase the biomass yield and decrease chemical variability of plants, but due to the removal of natural environmental factors, it may reduce secondary metabolite production (White et al., 2008). Plants can sustain growth under drought conditions through osmoregulation which occurs via accumulation of non- reducing sugars, reducing sugars, organic acids or amino acids (Büssis and Heineke, 1998). Coumarins are synthesized in plant leaves and stored in the roots; therefore increasing leaf growth through irrigation and nitrogen application, may increase the concentration of chemical compounds in roots. This study was aimed at establishing whether there is any relationship between soil water, nitrogen and the chemical composition of P. sidoides.

5.2. Materials and methods

For a detailed description of the field trial please refer to Chapter 3.

5.2.1. Thin layer chromatography (TLC)

The European Pharmacopoeia 6.0 (2008) was adapted for the TLC (aluminum, 60f254) assay. Standard compounds, scopoletin and esculin, were acquired from Sigma Aldrich (South Africa). To prepare a reference solution 1 mg of scopoletin and 1 mg of esculin, respectively, were dissolved in 1 ml of methanol. As for the test solution, 50 mg of the powdered P. sidoides root samples were dissolved in 8 ml methanol.

5.2.2. Metabolomic analysis The sample preparation, extraction, data acquisition, analysis, data mining and processing were performed by adapting the methods described by Maree and Viljoen (2012), Mediani et al. (2012) and Yang et al. (2012).

5.2.2.1.Sample preparation and extraction method The plants were harvested 12 months after planting and the roots were chopped and oven dried (Economy oven, 620 digital, Labotec) at 70 °C for 48 hours. A powdered sample of 50 mg per treatment was weighed in 2 ml Eppendorf tubes for extraction and analysis. Added to the samples was 0.75 mL of CH3OH-d4 and 0.75 mL of potassium dihydrogen phosphate

(KH2PO4), buffered in deuterium water (D2O) (pH 6.0) containing 0.1% (w/w) TSP (Trimethylsilylpropionic acid sodium salt). The mixture was vortexed at room temperature for 1 min, ultrasonicated for 20 min and then centrifuged for 20 minutes. The supernatant from each tube was transferred to an NMR tube for analysis.

5.2.2.2. Data acquisition/sample analysis NMR (nuclear magnetic resonance) spectral data were obtained using a 600 MHz 1HNMR spectrometer (Varian Inc, California, USA), with 32 scans recorded. The NMR gives information about the structure of a compound, by placing a substance in a strong magnetic field that affects the spin of its atomic nuclei. A radio wave passes through the substance, reorienting the nuclei; and as the wave is turned off the nuclei releases a pulse of energy or signals that provides data on the molecular structure of the substance (Anand et al. 2011). These signals are characterized by their frequency or chemical shift, intensity, fine structure and 72

magnetic relaxation properties, all of which reflects the environment of the detected nuclei (Krishna et al., 2005).

The phasing and baseline corrections were conducted using MestReNova software (9.0.1, Mestrelab Research, Spain), with consistent settings for all sample spectra. The chemical shift ranges of δ 4.70 – 4.90 and δ 3.23 – 3.36, representing water and methanol respectively (Mediani et al., 2012), were excluded from further analysis.

5.2.2.3. Data mining and processing MestReNova software (9.0.1, Mestrelab Research, Spain) was further used for bucketing of NMR spectra. The NMR regions were divided into 0.04 ppm bins resulting in 249 integrated regions.

Thereafter Multivariate data analysis (MVDA) was performed by principal component analysis (PCA), orthogonal partial least square discriminatory analysis (OPLS-DA) and hierarchical cluster analysis (HCA). This was performed with SIMCA-P software (13.0, Umetrics, Sweden) using the Parreto scaling method.

The PCA is an unsupervised analysis performed to provide an overview of all observation on the sample (Maree and Viljoen, 2011). Scatter score plots from the PC analysis were constructed to identify and evaluate groupings, trends and strong outliers (Maree and Viljoen, 2011). The second phase of analysis, the OPLS-DA, is a supervised pattern recognition method of which the main purpose is to separate the systematic variation in the X-matrix into two parts with one part linearly related to the Y-matrix and one that is unrelated to the Y-matrix (Maree and Viljoen, 2011).

5.3. Results

The thin layer chromatography (TLC) results showed the presence of the two standard compounds, scopoletin and esculin, in all the 12 treatments (Figure 5.1). Umckalin was excluded as it is only an indicator compound for P. sidoides. The TLC plate however did not seem to indicate any difference in the concentration of the compounds between the samples, as all the treatments showed similar absorbance of UV light. When comparing the standard

compounds (1mg/ml) with the compounds in the sample, it is evident that the compounds were found in very low concentrations in the root samples.

N: 0 50 100 150

% ADL: 30 50 70 30 50 70 Scop Esc 30 50 70 30 50 70

Figure 5.1. TLC analysis showing the presences of scopoletin (scop) and esculin (esc) in P. sidoides plants exposed to different water and nitrogen treatments. N = nitrogen in kg · ha-1, ADL = allowable depletion level of plant available water.

The results from a principal component analysis (PCA), which is an unsupervised analysis method, did not show good variation between data sets (Figure 5.2). There was no separation between the well watered (30% ADL, green), the moderately stressed (50% ADL, blue) and the severely stressed (70% ADL, red) treatments.

Figure 5.2. Principal component analysis (PCA) results of P. sidoides on NMR spectra, with the ellipse representing Hotelling within 95% confidence. The well watered treatments is represented by green colour, moderately stressed by blue colour and severely stressed by red colour.

However, when the data sets were exposed to a supervised method, orthogonal partial least square discriminatory analysis (OPLS-DA), there was a separation observed between the well watered treatment and the water stressed treatments (Figure 5.3). Accordingly the severely stressed treatments are generally clustered to the left of the ellipse, with the moderately stressed treatments more to the centre, indicating a gradual change from the well watered treatments on the right to the severely stressed treatments on the left. Two groups could be observed where the well watered treatment was separated from the two water stressed treatments. The hierarchical cluster analysis (Figure 5.4) further shows similar groupings.

Figure 5.3. Orthogonal partial least square discriminatory analysis (OPLS-DA) results of P. sidoides on NMR spectra, with the ellipse representing Hotelling within 95% confidence. The well watered treatments is represented by green colour, moderately stressed by blue colour and severely stressed by red colour.

Figure 5.4. Hierarchical clustering showing the two separated groups from results OPLS-DA of P. sidoides plants exposed to different water treatments. The green group represents mostly the well watered treatments with none of these well watered treatments in the blue group (water stressed). 76

Further investigations were performed to reveal the variants which contributed to the separation. Figure 5.5 (A) shows the variants (representing specific NMR regions) which contributed, positively and negatively, to the grouping of the well watered treatment, while Figure 5 (B) shows the variants for the water stressed treatments. The major compounds identified from the contribution plots are asparagine (1), sucrose (2), xylose (3), glucose (4), citric acid (5) and arginine (6), as shown in Table 5.1.

4 2 3 3

5 5

Average), Weight=p1

–

contribution (Group 1

6 Score Score

Variant ID (Primary)

Average), Weight=p1

–

5 5 4 2 3 3 B

contribution (Group Score Score

Variant ID (Primary) Figure 5.5. Score contribution plot showing NMR regions which had an effect on the separation of the well watered treatments (A) and water stressed treatments (B), in the OPLS-DA. 1 = asparagine, 2 = sucrose, 3= xylose, 4 = glucose, 5 = citric acid, 6 = arginine

Table 5.1. Specific NMR regions and compounds which contributed to the separation of the well watered treatment from the water stressed treatment.

Treatment Variant NMR value Compound Number

Well watered 32, 34 1.28 Unidentified -

54 1.36 Unidentified -

56 2.24 Unidentified -

57 2.28 Unidentified -

63 2.52 Citric acid 5

67 2.68 Citric acid 5

85 - 87 3.4 – 3.48 Glucose 4

89, 90 3.56, 3.6 Glucose 4

91 - 92 3.64 – 3.68 Glucose 4

94 – 98 3.76 – 3.92 Glucose 4

101 – 102 4.04 – 4.08 Sucrose 2

104 – 105 4.16 – 4.7 Sucrose 2

115 4.55 Xylose 3

130 5.15 Xylose 3

136 5.44 Sucrose 2

148 5.92 Unidentified -

149 5.96 Unidentified -

166 6.64 Unidentified -

Water stress 69 - 74 2.78 – 2.95 Asparagine 1

99 3.95 Unidentified -

41 - 43 1.64 – 1.72 Arginine 6

46 – 47 1.84 – 1.88 Unidentified -

57 2.28 Unidentified -

67 2.68 Unidentified -

Both the PCA (Figure 5.6) and the OPLS-DA analysis (Figure 5.7) on the nitrogen treatments did not show separation between the data sets.

Figure 5.6. Principal component analysis (PCA) results of P. sidoides on NMR spectra, with the ellipse representing hotelling within 95% confidence. The N treatments are represented as follows: no nitrogen (green), 50 (blue), 100 (red) and 150 (yellow) kg · N · ha-1.

Figure 5.7. Orthogonal partial least square discriminatory analysis (OPLS-DA) results of P. sidoides on NMR spectra, with the ellipse representing hotelling within 95% confidence. The Nitrogen treatments are represented as follows: no nitrogen (green), 50 (blue), 100 (red) and 150 (yellow) kg · N · ha-1.

5.4. Discussion

The PCA analysis did not show any separation for water treatments as the data sets were scattered all over in the ellipse. However when the data sets were transformed to the OPLS- DA analysis there were groupings based on largest variation between the samples to separate treatments. Nitrogen did not show any effect on variation on both the PCA and OPLS-DA analysis methods, which is indicative of a very similar metabolite profile with the application of different nitrogen levels.

Although the two main compounds, scopoletin and esculin, were observed for TLC analysis they were not detected for NMR. This could be because both scopoletin and esculin were present at low concentrations which were below the sensitivity threshold (Krishna et al., 2005) of the NMR (Shulaev et al., 2008). Yang et al. (2012) mentioned that the content of some compounds increases with increasing age of plants, which could explain the low concentrations of the two compounds. In the present study plants were harvested at the age of one year, while harvesters are encouraged to harvest plants that are at least two years old. It seemed however that the two compounds, scopoletin and esculin, did not differ significantly between the treatments and other compounds in higher concentrations were found to be responsible for the separation between the treatments.

The compounds that contributed to the groupings were citric acid, glucose, sucrose and xylose for the well watered treatment, while for the water stressed treatment it was mainly asparagine and arginine. The well watered treatment showed an increase in primary metabolites, especially the sugars. As for the water stressed treatments the amino acids, asparagine and arginine increased.

Stress induces secondary metabolite synthesis in preparation for defense, whereas under no stress primary metabolites will be synthesized, for growth. Accumulation of compounds such as free amino acids, under drought conditions, ensure osmotic adjustment of plants (Simon- Sarkadi et al., 2006 and Lea et al., 2007). Singh et al. (1973) reported an increase in concentrations of asparagine, amongst other free amino acids, in the leaves of barley plants exposed to water stress. In potato leaves, no significant differences were observed in sucrose content between the control and water stress treatment, which was induced by application of polyethylene glycol. However, there was a significant increase in the amino acid content in the 81

old leaves of water stressed plants, seven days after inducing the stress (Büssis and Heineke, 1998). This increase was attributed to an increase in asparagine, amongst other amides. Fukutoku and Yamada (1981) also reported a significant increase in amino acid content, especially proline and asparagine, in water stressed soybean plants compared to non-stressed plants. Sucrose and glucose content were also reported to decrease in the leaves and roots of Lupinus albus plants exposed to water stress (Pinheiro et al., 2004). Furthermore, asparagine was decreased in the young leaves and roots of L. albus due to water stress, though in the stem it was significantly increased (Pinheiro et al., 2004).

5.5. Conclusions The study successfully differentiated the metabolites responsible for variation between well watered and water stressed root samples of P. sidoides, using NMR. This was a quick and easy method applied to obtain a profile of all metabolites present in the samples. The OPLS-DA allowed the separation of the three irrigation treatments into two clusters, the well watered and water stressed (moderately and severely stressed) clusters. However, there was no good separation regarding the nitrogen treatments.

The primary metabolites identified in the well watered treatment were mostly sugars and in the water stressed treatments, amino acids asparagine and arginine. Secondary metabolism, finely tuned, may not be relevant for water stress levels as plants need to survive first. A consideration may be to grow the plants under well watered conditions for a year or two to increase growth and root yield. Thereafter plants can be exposed to water stress in the second season, which is expected to increase the concentration of important secondary metabolites. The lack of information on the compounds responsible for the medicinal activity of P. sidoides however poses a challenge in targeting specific compounds in secondary metabolism.

Liquid chromatography-Mass spectrometry (LC-MS) or Gas chromatography-MS (GC-MS) techniques may be more appropriate for identifying and quantifying lower concentration compounds as they are more sensitive than the NMR and should be considered in future studies.

References

Alvarenga, I.C.A., Valadares, R.V., Martins, E.R., Oliveira, F.G., De Figueiredo, L.S. and Kobayashi, M.K. 2011. Water stress before harvest of pepper-rosmarin. Pesq. Agropec. Bras. 46: 706 – 711

Araya, H.T., Soundy, P., Steyn, J.M. 2008. Response of herbage yield, essential oil yield and composition of South African rose-scented geranium (Pelargonium sp.) to conventional and organic nitrogen. J. Essent. Oil Res. 18: 111-115

Bladt, S. and Wagner, H. 2007. From the Zulu medicine to the European phytomedicine Umckaloabo®. Phytomedicine 14: 2 – 4

Bourgaud, F., Hehn, A., Larbat, R., Doerper, S., Gontier, E., Kellner, S. and Matern, U. 2006. Biosynthesis of coumarins in plants: a major pathway still to be unravelled for cytochrome P450 enzymes. Phytochem. Rev. 5: 293 – 308

Brendler, T., van Wyk, B.-E. 2008. A historical, scientific and commercial perspective on the use of Pelargonium sidoides (Geraniaceae). J. Ethnopharmacol. 119: 420 – 433

Büssis, D. and Heineke, D. 1998. Acclimation of potato plants to polyethylene glycol- induced water deficit II. Contents and subcellular distribution of organic solutes. J. Exp. Bot. 49: 1361 – 1370

European medicines agency, 2012. Assessment report on Pelargonium sidoides DC. and/or Pelargonium reniforme Curt., radix. EMA/HMPC/560962/2010. www.ema.europa.eu Accessed on 31st October 2010

European pharmacopoeia 6.0. 2008. Pelargonium root. 2625. www.edqm.eu. Accessed on 31st October 2010

Fukutoku, Y. and Yamada, Y. 1981. Diurnal changes in water potential and free amino acid content of water-stressed and non-stressed soybean plants. Soil Sci. Plant Nutr. 27: 195 – 204 83

Gödecke, T., Kaloga, M. and Kolodziej, H. 2005. A phenol glucoside, uncommon coumarins and flavonoids from Pelargonium sidoides DC. Z. Naturforsch. 60b: 677 – 682

Jain, P.K. and Joshi, H. 2012. Coumarin: chemical and pharmacological profile. J. Appl. Pharma. Sci. 6: 236 – 240

Kayser, O. and Kolodziej, H. 1995. Highly oxygenated coumarins from Pelargonium sidoides. Phytochemistry 5: 1181 – 1185

Kolodziej, H. 2002. Pelargonium reniforme and Pelargonium sidoides: their botany, chemistry and medicinal use. In: Lis-Balchin, M. (ed), Geranium and Pelargonium: The genera Geranium and Pelargonium. Taylor and Francis, London, pp 262 – 290

Krishna, P., Kruger, N.J. and Ratcliffe, R.G. 2005. Metabolite fingerprinting and profiling in plants using NMR. J. Exp. Bot. 56: 255 – 265

Lacy, A. and O’Kennedy, R. 2004. Studies on coumarins and coumarin-related compounds to determine their therapeutic role in the treatment of cancer. Curr. Pharm. Design 10: 3797 – 3811

Lea, P.J., Sodek, l., Parry, M.A.J., Shewry, P.R. and Halford, N.G. 2007. Asparagine in plants. Ann. Appl. Biol. 150: 1 – 26

Maree, J.E. and Viljoen, A.M. 2012. Phytochemical distinction between Pelargonium sidoides and Pelargonium reniforme – A quality control perspective. S. Afric. J. Bot. 82: 83 – 91

Mediani, A., Abas, F., Khatib, A., Maulidiani, H., Shaari, K., Choi, Y.H. and Lajis, N.H. 2012. 1H-NMR based metabolomics approach to understanding the drying effects on the phytochemicals in Cosmos caudatus. Food Res. Int. 49: 763 – 770

Ojala, T. 2001. Biological screening of plant coumarins. PhD thesis. University of Helsinki. Finland.

Pinheiro, C., Passarinho, J.A. and Ricardo, C.P. 2004. Effect of drought and rewatering on the metabolism of Lupinus albus organs. J. Plant Physiol. 161: 1203 – 1210

Pirzad, A., Khoshbakht, M., Siadat, A.A., Fathi, G.A. and Bakhshandeh, A.M. 2012. Growth analysis of Pimpinella anisum under different irrigation regimes and amounts of super absorbent polymer. Intl. Res. J. Appl. Basic Sci. 3: 112 – 122

Seidel, V. and Taylor, P.W. 2004. In vitro activity of extracts and constituents of Pelagonium against rapidly growing mycobacteria. Int. J. Antimicrob. Ag. 23: 613 – 619

Shenkut, A., Tesfaye, K. and Abegaz, F. 2013. Determination of water requirement and crop coefficient for sorghum (Sorghum bicolor L.) at Melkassa, Ethopia. Sci. Technol. Arts Res. J. 2: 16 – 24

Shulaev, V., Cortes, D., Miller, G. and Mittler, R. 2008. Metabolomics for plant stress response. Physiol. Plant. 132: 199 – 208

Singh, T.N., Paleg, L.G. and Aspinall, D. 1973. Stress metabolism I: Nitrogen metabolism and growth in the barley plant during water stress. Aust. J. Biol. Sci. 26: 45 – 46

Solecka, D. 1997. Role of phenylpropanoid compounds in plant response to different stress factors. Acta Physiol. Plant.19: 257 – 268

Simon-Sakardi, L., Kocsy, G., Várhegyi, Á., Galiba, G. and De Ronde, J.A. 2006. Sress induced changes in the free amino acid composition in transgenic soybean plants having increased proline content. Biol. Plantarum 50: 793 – 796

Tugizama, F., Piater, L and Dubery, I. 2013. Plant metabolomics: A new frontier in phytochemical analysis. S. Afric. J. Sci. 109, http://dx.doi.org/10.1590/sajs. 2013/20120005

Van Wyk, B.-E. 2008. A broad review of commercially important southern African medicinal plants. J. Ethnopharmacol. 119: 342 – 355

Wittschier, N., Faller, G. and Hensel, H. 2007. An extract of Pelargonium sidoides (EPs 7630) inhibits in situ adhesion of Helicobacter pylori to human stomach. Phytomedicine 14: 285–288

Yang, S.-O., Shin, Y.-S, Hyun, S, -H, Cho, S., Bang, K.-H., Lee, D., Choi, S.P. and Choi, H. -K. 2012. NMR-based metabolic profiling and differentiation of ginseng roots according to cultivation ages. J. Pharm. Biomed. Anal. 58: 19 – 24

CHAPTER 6

GENERAL DISCUSSION, CONCLUSIONS AND RECOMMENDATIONS

When irrigation must supplement rainfall for crop production, it results in competition between domestic, industrial and agricultural water users. The need for a better understanding of both crop water requirements and its relationship with yield becomes more important in optimization of water application to crops. Thus the determination of exactly when is a shortage or excess of water causing production losses, important for appropriate irrigation management. Irrigation management and nutrient management, especially nitrogen, indirectly affect each other. Under drought conditions root activity is constrained, leading to reduced nitrogen uptake. Nitrogen on the other hand increases vegetative growth, which has an effect on water uptake through an increased transpiration area. The yield and composition of secondary metabolites within a species may vary between plants from different geographical locations and may be influenced by environmental stresses, amongst other factors, and their intensities.

The results from this study showed a relationship between physiological, morphological and yield response as well as chemical composition of P. sidoides. The observed stomatal closure under water stress conditions, due to low turgor pressure in the guard cells, had a significant negative effect on leaf area, leaf area index, total biomass and root yield. All the above mentioned parameters decreased with water stress, although there were not always significant differences between moderately stressed and severely stressed treatments. Regulation of stomatal conductance is one of the main mechanisms of reducing water loss under drought conditions, however the uptake of primary substrates for photosynthesis is also compromised. Another response to stress is the reduction in leaf area and leaf number. Any reduction in these further leads to a reduced photosynthetic area. All this resulted in a significant decrease in total biomass and root yield from the well watered treatment to the water stressed treatments.

The chlorophyll content, biomass nitrogen content and nitrogen uptake results showed a significant increase under water stress compared to the well watered treatment. The less turgid cells under water stress could mean that the chloroplasts are closer together, giving a higher chlorophyll content reading on the measured leaf area, while under well watered conditions the 87

open stomata resulted in spaces between the chloroplasts giving a low chlorophyll content reading. It has been reported that under water stress conditions, internal nitrogen concentrations rise due to concentration effect, and this could be because plant growth and development are reduced, thus the build-up of nitrogen in the plant.

The results further showed an increase in the nitrogen use efficiency and while water use efficiency decreased, under well watered conditions. Though the nitrogen uptake was low in well watered treatments, it was efficiently used for growth and development. The significantly increased yields under well watered conditions also resulted in a significantly higher water use efficiency under those conditions.

There was a significant positive effect of nitrogen on number of leaves, but not on leaf area, at the application rate of 100 kg · N · ha-1. The total biomass yield increased significantly with application of nitrogen, compared to the control, with no significance differences between N levels. However, a trend of a decrease in total biomass above 150 kg · N · ha-1 was observed and this could indicate a possibility of luxury consumption beyond 150 kg · N · ha-1. The lack of a significant effect of 100 kg · N · ha-1on leaf area and total biomass, which was observed on number of leaves, could possibly mean there was stimulation of a large number of small leaves. Furthermore, nitrogen significantly increased nitrogen content, of the whole plant. This can be linked to the increase in total biomass, with increasing nitrogen level. Nitrogen is known to promote vegetative growth, but it did not translate into an increase in root yield in this study. This could have been due to sufficient residual N in the soil. The nitrogen use efficiency was significantly reduced by nitrogen application beyond 50 kg · N · ha-1and this could possibly emphasize the luxury consumption under increased nitrogen levels.

The investigation into chemical composition of roots showed that it was affected by water stress, but not by nitrogen. Asparagine content, which is an effective molecule for storage and transportation of nitrogen in living organisms, increased with water stress. This could be linked to the increases in nitrogen and chlorophyll contents which was observed under water stress. The increases in total biomass and root yield under well watered conditions could be attributed to the increased primary metabolite content, under such conditions.

Regarding our hypotheses, the following can be concluded:

i. There is an optimum N level (e.g. 100 kg · N · ha-1) that will improve growth and increase yield of storage roots with no negative effect on medicinal active compound levels of the plant.

This hypothesis is rejected as nitrogen level did not have any significant effect on root yield.

ii. Improved root yield will lead to increases in total yield of the medicinally active compounds.

This hypothesis is also rejected as nitrogen application did not have any significant effect on root yield. It is however not yet clear if nitrogen application increased the total yield of medicinally active compounds.

iii. Increasing nitrogen rates will significantly increase chlorophyll content and leaf area.

This hypothesis is also rejected as nitrogen application only had a significant effect on number of leaves.

iv. Appropriate irrigation scheduling will improve yield of storage roots with no negative effect on active compounds, and thus increase the total yield of active compounds.

The hypothesis is accepted as irrigating at 30% ADL of PAW resulted in increased root yield which translate to increased total yield of active compounds.

From this study it can be recommended that further investigations should be conducted, with more sensitive techniques such as Liquid chromatography-Mass spectrometry (LC-MS) or Gas chromatography-MS, to quantify the content of umckalin, scopoletin and esculin. The presence of these compounds has however not been directly linked to the medicinal use of P. sidoides although they are generally used as marker compounds. From the NMR metabolomics profile it would seem as if there are also other compounds in high concentrations that could also contribute to the medicinal activity of the plants and therefore more information is needed to conclusively decide if water and nitrogen is affecting the medicinal content and quality of the

plant material. What is important to note is that the major differences observed between the treatments are due to primary metabolites present and that the role of these compounds might be overlooked in the search for secondary metabolites, which might be present in very low concentrations. It is evident that the plant must ensure survival first after a change in environmental conditions, which can be seen in the major change in primary metabolite content of well watered treatment compared to the water stressed treatments. The increase in primary metabolites also shows that the plant is growing better than in the water stressed conditions. The general perception that secondary metabolism is affected by environmental conditions is not supported by the evaluation of the two marker compounds in the study. The compounds are generally accepted to be present in the root samples of P. sidoides and P. reniforme, but the actual contribution to the medicinal activity and therefore quality of the extract has not been confirmed yet. The information can then be used to confirm if the reduced root yield under water stress is not compensated for by an increase in content of the mentioned active compounds and only then can recommendations be made to farmers. Further research under different soil conditions and other nutrient stress factors should also be investigated. A further recommendation is that plants should be grown for two years or more before harvesting to ensure higher levels of secondary metabolites when stressed.

Appendix 1

Randomisation of the trial, with the blocks representing replicates. N = nitrogen in kg · ha-1 and allowable depletion level (ADL) as percentage (%) of plant available water.

File name is Pelargonium sidoides randomisation.gen

To determine the effect of different N levels and soil water depletion regimes on yield and chemical composition of P. sidoides Randomised block design 3 sets (blocks or reps) ======of 12 random numbers for 4 N x 3 ADL

TMTCOMB N ADL 1 0 30 2 0 50 3 0 70 4 50 30 5 50 50 6 50 70 7 100 30 8 100 50 9 100 70 10 150 30 11 150 50 12 150 70

===== BLOCK 1 =====

POSITION TMTCOMB 1 7 2 4 3 1 4 10 5 3 6 11 7 12 8 6 9 2 10 5 11 8 12 9

===== BLOCK 2 =====

POSITION TMTCOMB 1 4 2 6 3 10 4 12 5 3 6 8 7 5 8 9 9 1 10 2 11 11 12 7

===== BLOCK 3 =====

POSITION TMTCOMB 1 6 2 10 3 3 4 2 5 12 6 11 7 5 8 1 9 9 10 4 11 7 12 8

End of Meshack Mofokeng - ARC-VOPI - Project no a070202b. Current data space: 1 block, peak usage 1% at line 33.

Appendix 2

The calibration curves that were used to determine the soil water loss at 0 - 0.2 (A), 0.2 - 0.4 (B), 0.4 - 0.6 (C), 0.6 - 0.8 (D) and 0.8 - 1.0 m (E) of the soil profile.

0.25 y = 0.0013x + 0.0008 R² = 0.9533

0.20 3

0.15 /cm

3 0.10 cm 0.05 A 0.00 0 50 100 150 200 CPS

0.3 y = 0.0011x - 0.1958 0.25 R² = 0.955

3 0.2

/cm 0.15 3

cm 0.1 0.05 B 0 0 200 400 600 CPS

0.3 y = 0.0013x - 0.3957 0.25 R² = 0.9868

3 0.2

/cm 0.15 3

cm 0.1 0.05 C 0 0 200 400 600 CPS

0.3 y = 0.0013x - 0.4005 0.25 R² = 0.9979

3 0.2

/cm 0.15 3

cm 0.1 D 0.05 0 0 200 400 600 CPS

0.3 y = 0.0013x - 0.3969 R² = 0.9851 0.25

3 0.2

/cm 0.15 3

cm 0.1 E 0.05 0 0 200 400 600

CPS

Appendix 3

Plant identification dispatch list: SANBI (South African National Biodiversity Institute).

SANBI

Biodiversity for Life Private Bag X101, Pretoria, 0001

Pretoria National Herbarium (PRE) South Africa

South African National Biodiversity Institute Tel: +27 12 8435000, Fax: +27 12 8043211

140º² Plant Identification Dispatch List 17 March 2013

Client Meshack Mofokeng

Address P/BAG X 293 Tel: 012-841-9611

PRETORIA Cell:

Email: [email protected]

ID CODES: 1 = Specimen too poor to ID 5 = Genus requiring/under revision

2 = Label information inadequate 6 = Specimen closest to name listed (cf) 3 = Cannot match specimen in herbarium 7 = Please send more material

4 = Specialist not available to do ID 8 = Please refer to attached note/letter

9 = New record

Collector No. Plant Name Det. By Det. Notes ID Code M, Mofokeng P.fi. Pelargonium sidoides /.DC’ Masupa. T.T. I7/3 /2013 0

Please note a handling fee is charged for each specimen received for identification 1

Appendix 4

Preparation of treatment plots under rainshelter (4.a and b), the trial after planting (4.c) and P. sidoides plants under well watered (4.d), moderately stressed (4.e) and severely stress conditions (4.f).

4.a 4.b

4.c

4.d 4.e 4.f